Guide to Adjusting Vision Parameters for Ordered Quadrilateral Loading and Unloading
This article mainly introduces how to adjust vision parameters according to the actual scenario in ordered loading and unloading and random picking of quadrilateral workpieces.
1. 2D Recognition
1.1 Preprocessing
The preprocessing for 2D recognition handles the 2D image before Instance Segmentation.
1.1.1 Bilateral Filtering
- Function
Image smoothing based on bilateral filtering
- Parameter description
| Parameter | Description | Default Value | Range |
|---|---|---|---|
| Maximum Depth Difference | The maximum depth difference for bilateral filtering | 0.03 | [0.01, 1] |
| Filter Kernel Size | The convolution kernel size for bilateral filtering | 7 | [1, 3000] |
1.1.2 Depth-to-Normal Map
- Function
Calculates pixel normals from the depth map and converts the image into a Normal map
1.1.3 Image Enhancement
- Function
Common image enhancement methods, such as saturation, contrast, brightness, and sharpness
- Parameter description
| Parameter | Description | Default Value | Range |
|---|---|---|---|
| Image Enhancement Type | Enhances a specific element of the image | Contrast | Saturation, contrast, brightness, sharpness |
| Image Enhancement Threshold | The degree to which a specific image element is enhanced | 1.5 | [0.1, 100] |
1.1.4 Histogram Equalization
- Function
Improves image contrast
- Parameter description
| Parameter | Description | Default Value | Range |
|---|---|---|---|
| Local Mode | Local or global histogram equalization. If selected, local histogram equalization is used; if cleared, global histogram equalization is used | Selected | / |
| Contrast Threshold | Contrast threshold | 3 | [1,1000] |
1.1.5 Filter Depth Map by Color
- Function
Filters the depth map according to color values
- Parameter description
| Parameter | Description | Default Value | Range |
|---|---|---|---|
| Fill Kernel Size | The size for color filling | 3 | [1,99] |
| Filter Depth by hsv - Maximum Color Range Value | Maximum color value | [180,255,255] | [[0,0,0],[255,255,255]] |
| Filter Depth by hsv - Minimum Color Range Value | Minimum color value | [0,0,0] | [[0,0,0],[255,255,255]] |
| Keep Regions Within the Color Range | If selected, regions within the color range are kept; if cleared, regions outside the color range are kept | / | / |
1.1.6 Gamma Image Correction
- Function
gamma correction changes image brightness
- Parameter description
| Parameter | Description | Default Value | Range |
|---|---|---|---|
| Gamma Compensation Coefficient | If this value is less than 1, the image becomes darker; if this value is greater than 1, the image variable | 1 | [0.1,100] |
| Gamma Correction Coefficient | If this value is less than 1, the image becomes darker and is suitable for overly bright images; if this value is greater than 1, the image becomes brighter and is suitable for overly dark images | 2.2 | [0.1,100] |
1.1.7 Fill Holes in the Depth Map
- Function
Fills holes in the depth map and smooths the filled depth map
- Applicable scenario
Due to issues such as occlusion from the workpiece structure itself and uneven lighting, the depth map may miss parts of the workpiece
- Parameter description
| Parameter | Description | Default Value | Range |
|---|---|---|---|
| Fill Kernel Size | The size for hole filling | 3 | [1,99] |
Fill Kernel Size can only be an odd number
- Parameter tuning
Adjust based on the detection result. If overfilling occurs, reduce the parameter; if filling is insufficient, increase the parameter.
- Example
1.1.8 Edge Enhancement
- Function
Sets edge areas of textures in the image to the Background color or to a color that differs significantly from the Background color, so as to highlight the edge information of the workpiece
- Applicable scenario
Edges are unclear due to workpieces occluding or overlapping each other
- Parameter description
| Parameter | Description | Default Value | Parameter Range | Tuning Suggestion |
|---|---|---|---|---|
| Normal Z-direction Filter Threshold | The angle filter threshold between the normal vector of each point in the depth map and the positive Z-axis of the Camera coordinate system. If the angle between the point normal and the positive Z-axis of the Camera coordinate system is greater than this threshold, the color at the corresponding position in the 2D image will be set to the Background color or to a color that differs significantly from the Background color | 30 | [0,180] | For flat workpiece surfaces, this threshold can be smaller. For curved workpieces, increase it appropriately according to the degree of surface inclination |
| Background | The RGB color threshold of the Background | 128 | [0,255] | / |
| Automatically Adjust Contrast Background | Selected After Automatically Adjust Contrast Background is enabled, the color of points in the 2D image whose angles are greater than the filter threshold is set to a color that differs significantly from the Background color threshold not selected Automatically Adjust Contrast Background, the color of points in the 2D image whose angles are greater than the filter threshold is set to the color corresponding to the Background color | Not selected | / | / |
- Example
1.1.9 Extract Top-layer Texture
- Function
Extracts the texture of the top layer or bottom layer workpiece, while setting other areas to the Background color or to a color that differs significantly from the Background color.
- Applicable scenario
Factors such as poor lighting conditions, similar color textures, tight stacking, interleaved stacking, or occlusion may make it difficult for the model to distinguish texture differences between upper and lower layer workpieces, which can easily lead to false detections.
- Parameter description
| Parameter | Description | Default Value | Parameter Range | Unit | Tuning Suggestion |
|---|---|---|---|---|---|
| Distance Threshold (mm) | If the distance between a point and the top-layer plane (bottom-layer plane) is lower than this threshold, the point is considered to be within the top-layer plane (bottom-layer plane) and should be retained; otherwise, it is considered a lower-layer (upper-layer) point, and the color of the lower-layer (upper-layer) point is set to the Background color or to a color that differs significantly from the Background color | 50 | [0.1, 1000] | mm | Generally set to 1/2 of the workpiece height |
| Clustered Point Cloud Quantity | The expected number of points participating in clustering, that is, the number of sampled Point Cloud points within the ROI 3D region | 10000 | [1，10000000] | / | The greater the Clustered Point Cloud Quantity, the slower the model Inference speed but the higher the accuracy; the smaller the Clustered Point Cloud Quantity, the faster the model Inference speed but the lower the accuracy |
| Minimum Number of Class Points | The minimum number of points used to filter classes | 1000 | [1, 10000000] | / | / |
| Automatically Calculate Contrast Background | Selected After Automatically Calculate Contrast Background is enabled, areas other than the top layer (bottom layer) in the 2D image are set to a color that differs significantly from the Background color threshold not selected Automatically Calculate Contrast Background, areas other than the top layer (bottom layer) in the 2D image are set to the color corresponding to the Background color threshold | Selected | / | / | / |
| Background Color Threshold | The RGB color threshold of the Background | 128 | [0,255] | / | / |
- Example
1.1.10 Remove the Image Background Outside roi3d
- Function
Removes the background outside the ROI3D region in the 2D image
- Applicable scenario
Excessive image background noise affects detection results
- Parameter description
| Parameter Name | Description | Default Value | Range |
|---|---|---|---|
| Fill Kernel Size | The size for hole filling | 5 | [1,99] |
| Number of Iterations | The number of Dilation iterations for the image | 1 | [1,99] |
| Automatically Calculate Contrast Background | Selected After Automatically Calculate Contrast Background is enabled, the area outside roi in the 2D image is set to a color that differs significantly from the Background color threshold not selected Automatically Calculate Contrast Background, the area outside roi in the 2D image is set to the color corresponding to the Background color threshold | Selected | / |
| Background Color Threshold | The RGB color threshold of the Background | 128 | [0,255] |
Fill Kernel Size can only be an odd number
- Parameter tuning
If more background noise needs to be removed from the image, reduce the Fill Kernel Size
- Example
1.2 Instance Segmentation
1.2.1 Scaling Ratio
- Function
Improves the accuracy and recall of 2D recognition by scaling the original image proportionally before Inference.
- Applicable scenario
If the detection effect is poor (for example, instances are not detected, missed, a bounding box covers multiple instances, or a bounding box does not fully cover an instance), this function should be adjusted.
Parameter description
Default Value: 1.0
Range: [0.01, 3.00]
Step size: 0.01
Parameter tuning
- Run with the default value and view the detection result in the visualization window. If instances are not detected, missed, a bounding box covers multiple instances, or a bounding box does not fully cover an instance, adjust this function.
In 2D recognition, the percentage shown for an instance is the Confidence score, and the number is the Instance ID (the recognition order of the instance).
In 2D recognition, the colored shadow on an instance is the Mask, and the rectangle surrounding the instance is the bounding box.
- Try different scaling ratios, observe changes in the detection results, and gradually determine the range of scaling ratios. If the detection effect improves significantly at a certain scaling ratio, use that scaling ratio as the lower bound; if the detection effect drops significantly at a certain scaling ratio, use that scaling ratio as the upper bound.
If good detection results still cannot be obtained after trying all scaling ratios, you may adjust the ROI region
As shown in the figure below, when the scaling ratio is 0.2, the detection effect improves significantly, so 0.2 can be determined as the lower bound of the scaling ratio range
When the scaling ratio is 0.5, the detection effect drops significantly, so 0.5 can be determined as the upper bound of the scaling ratio range
- If the actual scenario does not require high picking accuracy, you can choose a scaling ratio with better detection performance within the [0.2,0.5) interval; if the actual scenario requires high picking accuracy, you should further refine the scaling ratio range and adjust it with a smaller step size until you find the scaling ratio with the best detection effect.
1.2.2 Lower Confidence Threshold
- Function
Retains only recognition results from the Deep Learning model whose scores are higher than the lower Confidence threshold
- Applicable scenario
When the instances selected by the detection result are not as expected, this function can be adjusted
- Parameter description
Default Value: 0.5
Range: [0.01, 1.00]
Parameter tuning
If the model detects too few instances, reduce this threshold. If the value is too small, it may affect the accuracy of image recognition.
If an overly small lower Confidence threshold causes incorrect instances to be detected and these incorrect instances need to be removed, increase this threshold. If the value is too large, the number of retained detection results may be zero, resulting in no output.
1.2.3 Enable Auto Enhancement
- Function
Combines all input scaling ratios and rotation angles for Inference and returns all combined results greater than the specified lower Confidence threshold, which can improve model Inference accuracy, but will increase processing time.
- Applicable scenario
A single scaling ratio cannot meet the requirements of the actual scenario, causing incomplete detection, or the object placement tilt is relatively large.
- Example
If Auto Enhancement - Scaling Ratio is set to [0.8, 0.9, 1.0], and Auto Enhancement - Rotation Angle is set to [0, 90.0] , then the values in the scaling ratios and rotation angles are combined pairwise. The model internally automatically generates 6 images for Inference, and finally merges these 6 Inference results together, outputting the results greater than the lower Confidence threshold.
Auto Enhancement - Scaling Ratio
- Function
Scales the original image multiple times and performs multiple rounds of Inference, outputting a consolidated Inference result
- Applicable scenario
A single scaling ratio cannot meet the requirements of the actual scenario, resulting in incomplete detection
- Parameter description
Default Value: [1.0]
Range: the range for each scaling ratio is [0.1, 3.0]
Multiple scaling ratios can be set, separated by English commas
- Parameter tuning
Enter multiple scaling ratios from 1.2.1 Scaling Ratio that produced good detection results
Auto Enhancement - Rotation Angle
- Function
Rotates the original image multiple times and performs multiple rounds of Inference, outputting a consolidated Inference result
- Applicable scenario
Use when the object placement deviates significantly from the coordinate axes
- Parameter description
Default Value: [0.0]
Range: the range of each rotation angle is [0, 360]
Multiple rotation angles can be set, separated by English commas
- Parameter tuning
Adjust Auto Enhancement - Rotation Angle according to the object angles in the actual scenario. The tilt angle can be judged based on the pattern and bag opening shape of sacks, or the edges and brand logo of cartons
1.3 Point Cloud Generation
| Instance Point Cloud generation method | Mask method (after segmentation) | - | Generate Point Cloud using the segmented instance Mask |
| Bounding box method (after segmentation) | Bounding box scaling ratio (after segmentation) | Generate Point Cloud using the segmented instance bounding box | |
| Whether color is required when generating Point Cloud (after segmentation) | Whether the generated instance Point Cloud needs color attached | ||
| Mask method (after filtering) | - | Generate Point Cloud using the filtered instance Mask | |
| Bounding box method (after filtering) | Bounding box scaling ratio (after filtering) | Generate Point Cloud using the filtered instance bounding box | |
| Whether color is required when generating Point Cloud (after filtering) | Whether the generated instance Point Cloud needs color attached |
If acceleration is not required, there is no need to use the Instance Filtering function. Use Mask method (after segmentation) or Bounding box method (after segmentation) to generate the instance Point Cloud. You can view the generated instance Point Cloud in the Project storage folder \ProjectName\data\PickLight\HistoricalDataTimestamp\Builder\pose\input folder;
If acceleration is required, you can use the Instance Filtering function to filter instances, and use Mask method (after filtering) or Bounding box method (after filtering) to generate the instance Point Cloud. You can view the generated instance Point Cloud in the Project storage folder \ProjectName\data\PickLight\HistoricalDataTimestamp\Builder\pose\input folder.
1.4 Instance Filtering
1.4.1 Filter by Bounding Box Area
- Function overview
Filters according to the pixel area of the detected instance bounding box.
- Applicable scenario
Suitable for scenarios where instance bounding box areas differ greatly. By setting the upper and lower bounds of the bounding box area, noise in the image can be filtered out, improving image recognition accuracy and preventing noise from increasing the processing time of subsequent steps.
- Parameter description
| Parameter | Description | Default Value | Parameter Range | Unit |
|---|---|---|---|---|
| Minimum Area (pixels) | This parameter sets the minimum filter area for the bounding box. Instances whose bounding box area is lower than this value will be filtered out | 1 | [1, 10000000] | pixels |
| Maximum Area (pixels) | This parameter sets the maximum filter area for the bounding box. Instances whose bounding box area is higher than this value will be filtered out | 10000000 | [2, 10000000] | pixels |
- Example
Run with the default values. You can check the bounding box area of each instance in the log, as shown below.
Adjust **Minimum Area ** and Maximum Area according to the bounding box area of each instance. For example, set Minimum Area to 20000 and Maximum Area to 30000 to filter out instances whose pixel area is smaller than 20000 or larger than 30000. You can view the instance filtering process in the log.
1.4.2 Filter by Bounding Box Aspect Ratio
- Function overview
Instances whose bounding box aspect ratio is outside the specified range will be filtered out
- Applicable scenario
Suitable for scenarios where the aspect ratios of instance bounding boxes differ greatly
- Parameter description
| Parameter | Description | Default Value | Parameter Range |
|---|---|---|---|
| Minimum Aspect Ratio | The minimum bounding box aspect ratio. Instances whose bounding box aspect ratio is lower than this value will be filtered out | 0 | [0, 10000000] |
| Maximum Aspect Ratio | The maximum bounding box aspect ratio. Instances whose bounding box aspect ratio is higher than this value will be filtered out | 10000000 | [0, 10000000] |
| Use X/Y Axis Side Length as Aspect Ratio | By default, this option is not selected. The ratio of the longer side to the shorter side of the bounding box is used as the aspect ratio, which is suitable when the lengths of the long and short sides of the bounding box differ greatly; After selection, the ratio of the side length of the bounding box along the X-axis to that along the Y-axis in the pixel coordinate system is used as the aspect ratio, which is suitable when the long-side/short-side ratio of most normal instance bounding boxes is similar, but the ratio of the length along the X-axis to the length along the Y-axis differs greatly for some abnormally recognized instance bounding boxes. | Not selected | / |
1.4.3 Filter Instances by Class ID
- Function overview
Filters by instance class
- Applicable scenario
Suitable for scenarios where incoming materials contain multiple types of workpieces
- Parameter description
| Parameter | Description | Default Value |
|---|---|---|
| Retained Class IDs | Retain instances whose class IDs are in the list. Instances whose class IDs are not in the list will be filtered out | [0] |
- Example
1.4.4 Filter by Side Length of the Instance Point Cloud
- Function overview
Filters based on the long side and short side of the instance Point Cloud
- Applicable scenario
Suitable for scenarios where the distances of the instance Point Cloud along the x-axis or y-axis differ greatly. By setting the distance range of the instance Point Cloud, noise in the image can be filtered out, improving image recognition accuracy and preventing noise from increasing the processing time of subsequent steps.
- Parameter description
| Parameter | Description | Default Value | Parameter Range | Unit |
|---|---|---|---|---|
| Short Side Length Range (mm) | The side length range of the short side of the Point Cloud | [0, 10000] | [0, 10000] | mm |
| Long Side Length Range (mm) | The side length range of the long side of the Point Cloud | [0, 10000] | [0, 10000] | mm |
| Lower Edge Denoising Limit (%) | Extracts the lower percentage limit of the X/Y values (Camera coordinate system) in the instance Point Cloud, removing Point Cloud points outside the upper and lower limits to avoid noise affecting length calculation | 5 | [0, 100] | / |
| Upper Edge Denoising Limit (%) | Extracts the upper percentage limit of the X/Y values (Camera coordinate system) in the instance Point Cloud, removing Point Cloud points outside the upper and lower limits to avoid noise affecting length calculation | 95 | [0, 100] | / |
| Side Length Type | Filters by the long side and short side of the instance Point Cloud. Instances whose long-side or short-side lengths are out of range will be filtered out | Instance Point Cloud Short Side | Instance Point Cloud Short Side; Instance Point Cloud Long Side; Instance Point Cloud Long Side and Short Side | / |
- Example
1.4.5 Filter by Class ID Based on the Classifier
- Function overview
Filters instances by category ID based on the classifier. Instances not in the reference categories will be filtered out.
- Applicable scenario
In scenarios with multiple types of workpieces, the vision model may detect multiple types of workpieces, while the actual task may require only one specific category of workpiece. In this case, this function can be used to filter out unnecessary workpieces
- Parameter description
The default value is [0], which means that by default, instances whose category ID is 0 are retained. Instances whose category ID is not in the list will be filtered out.
1.4.6 Filter by Three-channel Color
- Function overview
Instances can be filtered out by three-channel color thresholds (HSV or RGB).
- Applicable scenario
Cases where incorrect instances and correct instances are clearly distinguishable by color.
- Parameter description
| Parameter | Description | Default Value | Range |
|---|---|---|---|
| Maximum Color Range Value | Maximum color value | [180,255,255] | [[0,0,0],[255,255,255]] |
| Minimum Color Range Value | Minimum color value | [0,0,0] | [[0,0,0],[255,255,255]] |
| Filter Percentage Threshold | Color pass-rate threshold | 0.05 | [0,1] |
| Reverse Filtering | If selected, instances with a proportion outside the color range lower than the threshold are removed; if not selected, instances with a proportion within the color range lower than the threshold in the instance image are removed | Not selected | / |
| Color Mode | The color space selected in color filtering | HSV color space | RGB color spaceHSV color space |
- Example
1.4.7 Filter by Confidence
- Function overview
Filters according to the Confidence score of the instance
- Applicable scenario
Suitable for scenarios where instance Confidence values differ greatly
- Parameter description
| Parameter | Description | Default Value | Parameter Range |
|---|---|---|---|
| Reference Confidence | Retains instances whose Confidence is greater than the threshold and filters out instances whose Confidence is less than the threshold. | 0.5 | [0,1] |
| Reverse Filtering Result | After reversal, retains instances whose visibility Confidence is less than the threshold and filters out instances whose Confidence is greater than the threshold. | Not selected | / |
- Example
1.4.8 Filter by Point Cloud Quantity
- Function overview
Filters according to the number of downsampled instance Point Cloud points
- Applicable scenario
The instance Point Cloud contains a large amount of noise
- Parameter description
| Parameter | Description | Default Value | Parameter Range |
|---|---|---|---|
| Minimum Point Cloud Quantity | The minimum value of the Point Cloud quantity | 3500 | [1, 10000000] |
| Maximum Point Cloud Quantity | The maximum value of the Point Cloud quantity | 8500 | [2, 10000000] |
| Filter Instances Whose Quantity Is Within the Interval | If selected, instances whose Point Cloud quantity is between the minimum and maximum values are filtered out; if not selected, instances whose Point Cloud quantity is not within the interval are filtered out | Not selected | / |
1.4.9 Filter by Mask Area
- Function overview
Filters image masks according to the sum of mask pixels (that is, the pixel area) of detected instances.
- Applicable scenario
Suitable for scenarios where instance Mask areas differ greatly. By setting the upper and lower bounds of the Mask area, noise in image masks can be filtered out, improving image recognition accuracy and preventing noise from increasing the processing time of subsequent steps.
- Parameter setting description
| Parameter Name | Description | Default Value | Parameter Range | Unit |
|---|---|---|---|---|
| Reference Minimum Area | This parameter is used to set the minimum filter area of the Mask. Instances whose Mask area is lower than this value will be filtered out | 1 | [1, 10000000] | pixels |
| Reference Maximum Area | This parameter is used to set the maximum filter area of the Mask. Instances whose Mask area is higher than this value will be filtered out | 10000000 | [2, 10000000] | pixels |
- Example
1.4.10 Filter by Visibility
- Function overview
Filters according to the visibility score of the instance
- Applicable scenario
Suitable for scenarios where instance visibility values differ greatly
- Parameter description
| Parameter | Description | Default Value | Parameter Range |
|---|---|---|---|
| Reference Visibility Threshold | Retains instances whose visibility is greater than the threshold and filters out instances whose visibility is less than the threshold. Visibility is used to determine how visible an instance in the image is. The more the workpiece is occluded, the lower the visibility. | 0.5 | [0,1] |
| Reverse Filtering Result | After reversal, retains instances whose visibility is less than the threshold and filters out instances whose visibility is greater than the threshold. | Not selected | / |
1.4.11 Filter Instances with Overlapping Bounding Boxes
- Function overview
Filters instances whose bounding boxes intersect and overlap
- Applicable scenario
Suitable for scenarios where instance bounding boxes intersect with each other
- Parameter description
| Parameter | Description | Default Value | Parameter Range |
|---|---|---|---|
| Bounding Box Overlap Ratio Threshold | The ratio threshold of the intersection area of bounding boxes to the area of the instance bounding box | 0.05 | [0, 1] |
| Filter Instances with Larger Bounding Box Area | If selected, the instance with the larger area among two instances with intersecting bounding boxes is filtered out; if not selected, the instance with the smaller area among two instances with intersecting bounding boxes is filtered out | Selected | / |
- Example
Newly added Filter enclosed instances. Run with the default values and check the intersection of instance bounding boxes in the log. After instance filtering, 2 instances remain.
The log shows that 12 instances were filtered out due to bounding box intersections, leaving 2 instances whose bounding boxes do not intersect
Set Bounding Box Overlap Ratio Threshold to 0.1 and select Whether to Filter Larger Instances. Check the instance filtering process in the log. 9 instances are filtered out because the ratio of the intersecting area of the bounding boxes to the area of the instance bounding box is greater than 0.1, 3 instances are retained because the ratio is less than 0.1, and 2 instances have non-intersecting bounding boxes.
Set Bounding Box Overlap Ratio Threshold to 0.1 and clear Whether to Filter Larger Instances. Check the instance filtering process in the log. The ratio of the intersecting area of the bounding boxes to the area of the instance bounding box is greater than 0.1 for 9 instances, but 2 of these instances are retained because their bounding box areas are smaller than the intersecting instances, so 7 instances are filtered out; 3 instances are retained because the ratio of the intersecting area of the bounding boxes to the area of the instance bounding box is less than 0.1, and 2 instances have non-intersecting bounding boxes.
1.4.12 [Master] Filter Instances with Concave or Convex Masks Based on the Area Ratio of Mask / Circumscribed Polygon of the Mask
- Function overview
Calculates the area ratio of Mask / circumscribed polygon of the Mask. If the ratio is lower than the set threshold, the instance will be filtered out
- Applicable scenario
Suitable for cases where the workpiece Mask has serrations / concave-convex irregularities.
- Parameter description
| Parameter | Description | Default Value | Range |
|---|---|---|---|
| Area Ratio Threshold | The threshold of the Mask / convex hull area ratio. If the ratio is lower than the set threshold, the instance will be filtered out. | 0.1 | [0,1] |
1.4.13 [Master] Filter by Average Point Cloud Distance
- Function overview
Filters based on the average distance from points in the Point Cloud to the fitted plane, removing uneven instance Point Cloud data
- Applicable scenario
Suitable for scenarios where planar workpiece Point Cloud data is bent
- Parameter description
| Parameter | Description | Default Value | Parameter Range | Unit |
|---|---|---|---|---|
| Plane Segmentation Distance Threshold (mm) | Extracts a plane from the bent instance Point Cloud. Points whose distance to the plane is less than this threshold are regarded as points on the plane | 10 | [-1000, 1000] | mm |
| Average Distance Threshold (mm) | The average distance from points in the instance Point Cloud to the extracted plane | 20 | [-1000, 1000] | mm |
| Remove Instances Whose Average Distance Is Less Than the Threshold | If selected, instances whose average distance from points to the extracted plane is less than the average distance threshold are filtered out; if not selected, instances whose average distance from points to the extracted plane is greater than the average distance threshold are filtered out | Not selected | / | / |
1.4.14 [Master] Filter Occluded Instances Based on the Area Ratio of Mask / Bounding Box
- Function overview
Calculates the area ratio of Mask / bounding box. Instances whose ratio is outside the maximum and minimum range will be filtered out
- Applicable scenario
Used to filter instances of occluded workpieces
- Parameter description
On the contrary, it indicates possible occlusion.
| Parameter | Description | Default Value | Range |
|---|---|---|---|
| Minimum Area Ratio | The lower bound of the Mask / bounding box area ratio range. The smaller the ratio, the more severely the instance is occluded | 0.1 | [0,1] |
| Maximum Area Ratio | The upper bound of the Mask / bounding box area ratio range. The closer the ratio is to 1, the lower the degree of occlusion of the instance | 1.0 | [0,1] |
1.4.15 [Master] Determine Whether All Top-layer Instances Have Been Fully Detected
- Function overview
One of the error-proofing mechanisms. It determines whether all top-layer instances have been detected. If there are top-layer instances that have not been detected, an error is reported and the Workflow ends.
- Applicable scenario
Suitable for scenarios where one image is used for multiple picks or picking must be performed in sequence, preventing missed picks in subsequent tasks caused by incomplete instance detection
- Parameter description
| Parameter | Description | Default Value | Parameter Range | Unit | Parameter Tuning |
|---|---|---|---|---|---|
| Distance Threshold | Used to determine top-layer workpieces. If the distance between a point and the highest point of the workpiece Point Cloud is less than the distance threshold, the point is considered a top-layer Point Cloud point; otherwise, it is not considered a top-layer Point Cloud point. | 5 | [0.1, 1000] | mm | Should be smaller than the height of the workpiece |
1.5 Instance Sorting
- Function overview
Groups, sorts, and extracts instances according to the selected strategy
- Applicable scenario
Applicable to depalletizing, random picking, and ordered loading and unloading scenarios
If sorting is not required, no specific strategy needs to be configured.
1.5.1 Reference Coordinate System
- Function overview
Sets a unified coordinate system for all instances to group and sort instances
- Applicable scenario
Applicable to depalletizing, random picking, and ordered loading and unloading scenarios
When using coordinate-related strategies, the Reference Coordinate System should be set first
- Parameter description
| Parameter | Description | Illustration |
|---|---|---|
| Camera Coordinate System | The coordinate system origin is above the object, and the positive Z-axis points downward; the XYZ values are the values of the object center point in this coordinate system |  |
| ROI Coordinate System | The coordinate system origin is roughly at the center of the stack, and the positive Z-axis points upward; the XYZ values are the values of the object center point in this coordinate system |  |
| Robot Coordinate System | The coordinate system origin is on the robot itself, and the positive Z-axis generally points upward; the XYZ values are the values of the object center point in this coordinate system |  |
| Pixel Coordinate System | The coordinate system origin is at the upper-left vertex of the RGB image and is a 2D planar coordinate system; X and Y are the x and y values of the bbox recognition box, and Z is 0 |  |
1.5.2 General Picking Strategy
- Parameter description
| Parameter | Description | Default Value |
|---|---|---|
| Strategy | Selects which value is used for grouping and sorting and how the sorting is performed, including the XYZ coordinate values of the instance Point Cloud center, bounding box aspect ratio, distance from the instance Point Cloud center to the ROI center, etc. Multiple rules can be stacked and executed sequentially | Instance Point Cloud center X-coordinate value from small to large (mm) |
| Grouping Step Size | According to the selected strategy, instances are divided into several groups based on the step size. The grouping step size is the interval between two groups of instances. For example, if the strategy selected is “Instance Point Cloud center Z-coordinate value from large to small (mm)”, then the Z coordinates of all instance Point Cloud centers are sorted from large to small, and then the Z coordinates are grouped according to the step size, with the corresponding instances also divided into several groups | / |
| Extract the First Several Groups | After grouping and sorting, specifies how many groups of instances need to be retained | 10000 |
| Strategy Name* | Description | Grouping Step Size | Extract the First Several Groups | |
|---|---|---|---|---|
| Default Value | Range | Default Value | ||
| Instance Point Cloud center XYZ coordinate values from large to small / from small to large (mm) | Uses the XYZ coordinate values of the Point Cloud center of each instance for grouping and sorting The Reference Coordinate System should be set first before using this strategy for sorting | 200.000 | (0, 10000000] | 10000 |
| From the middle to both sides / from both sides to the middle along the XY coordinate axes of the instance Point Cloud center (mm) | Uses the XY coordinate values of the Point Cloud center of each instance for grouping and sorting in the direction of “from the middle to both sides” or “from both sides to the middle” The Reference Coordinate System should be set first before using this strategy for sorting | 200.000 | (0, 10000000] | 10000 |
| Bounding box center XY coordinate values from large to small / from small to large (mm) | Uses the XY coordinate values of the center point of the bounding box of each instance in the pixel coordinate system for grouping and sorting | 200.000 | (0, 10000000] | 10000 |
| Bounding box aspect ratio from large to small / from small to large | Uses the ratio of the long side to the width side of the bounding box for grouping and sorting | 1 | (0, 10000] | 10000 |
| From the middle to both sides / from both sides to the middle along the XY coordinate axes of the bounding box center (mm) | Uses the XY coordinate values of the center point of the bounding box for grouping and sorting in the direction of “from the middle to both sides” or “from both sides to the middle” | 200.000 | (0, 10000000] | 10000 |
| Workpiece Type ID from large to small / from small to large | Uses the ID of the workpiece type for grouping and sorting, suitable for multi-type workpiece scenarios | 1 | [1, 10000] | 10000 |
| Local Feature ID from large to small / from small to large | Uses the ID of local features for grouping and sorting | 1 | [1, 10000] | 10000 |
| Confidence from large to small / from small to large | Uses the Confidence of each instance for grouping and sorting | 1 | (0, 1] | 10000 |
| Visibility from small to large / from large to small | Uses the visibility of each instance for grouping and sorting | 1 | (0, 0.1] | 10000 |
| Mask area from large to small / from small to large | Uses the Mask area of each instance for grouping and sorting | 10000 | [1, 10000000] | 10000 |
| Distance from instance Point Cloud center to ROI center from near to far / from far to near (mm) | Uses the distance between the Point Cloud center of each instance and the center of the ROI coordinate system for grouping and sorting | 200.000 | (0, 10000000] | 10000 |
| Distance from instance Point Cloud center to Robot coordinate origin from near to far / from far to near (mm) | Uses the distance between the Point Cloud center of each instance and the origin of the Robot coordinate system for grouping and sorting | 200.000 | (0, 10000000] | 10000 |
- Example
1.5.3 Custom Picking Strategy
(1) Function description
Switch Picking Strategy to Custom Picking Strategy, and click Add to add a custom picking strategy.
Customize the picking sequence for each workpiece. If it is difficult to achieve picking with the general picking strategy, or it is difficult to tune appropriate parameters because of issues such as Point Cloud noise, you can consider using a custom picking strategy
The custom picking strategy is suitable for depalletizing scenarios and ordered loading and unloading scenarios, but not for random picking scenarios, because the workpieces in a custom picking strategy must be ordered (that is, the sequence of workpieces is fixed)
A custom picking strategy can only be used in combination with a single general picking strategy, and the strategy can only select Z-coordinate from small to large
(2) Parameter description
| Parameter | Description | Default Value | Range | Parameter Tuning |
|---|---|---|---|---|
| IOU Threshold | Represents the overlap threshold between the annotated bbox and the detected bbox. The overlap is used to determine which image's sorting method should be selected when sorting the current workpiece instance. | 0.7 | [0,1] | The larger the threshold, the stricter the matching and the poorer the anti-interference capability. Minor shape or position changes may cause matching failure, which may match the wrong custom strategy and sort in the wrong order |
| Pixel Distance Threshold | Represents the size difference between a matchable bbox and the detected bbox. | 100 | [0,1000] | The smaller the threshold, the stricter the matching and the better the anti-interference capability. If the workpiece placement between different layers is relatively similar, the custom strategy may also be mismatched, resulting in an incorrect sorting order. |
(3) Select the Reference Coordinate System
When using a custom picking strategy, only the Camera coordinate system or the pixel coordinate system can be selected
If there are multiple layers of workpieces, select the Camera coordinate system; if there is only one layer of workpieces, select the pixel coordinate system
(4) Strategy, Grouping Step Size, Extract the First Several Groups
| Parameter | Description | Default Value |
|---|---|---|
| Strategy | Only Instance Point Cloud center Z-coordinate value from large to small / from small to large (mm) can be selected | / |
| Grouping Step Size | Based on the strategy of sorting Z coordinates from small to large, the Z coordinates of instances are sorted from small to large, and instances are divided into several groups according to the step size | 10000 |
| Extract the First Several Groups | After grouping and sorting, specifies how many groups of instances need to be retained | 10000 |
(5) Capture Image / Add Local Image
Click Capture Image to obtain an image from the currently connected Camera, or click Add Local Image to import an image from the local device. The number of images that need to be captured or added locally depends on the number of layers or the number of different placement forms of workpieces. If each layer is identical, only one image is required. Right-click the image to delete it.
On the acquired image, click and hold the left mouse button and drag to annotate the bbox. The DELETE key can be used to progressively delete annotated bbox boxes.
2. 3D Calculation
2.1 Preprocessing
The preprocessing for 3D calculation processes the 3D Point Cloud before pose estimation and Pick Point generation for instances
2.1.1 Point Cloud Cluster Denoising
- Function
Removes noise through Point Cloud clustering
- Applicable scenario
There is a large amount of noise in the instance Point Cloud
- Parameter description
| Parameter Name | Description | Default Value | Range | Unit | Tuning Suggestion |
|---|---|---|---|---|---|
| Distance Threshold for Point Cloud Clustering (mm) | Determines whether Point Cloud points in space belong to the same category. If the distance between Point Cloud points is lower than this threshold, they are considered the same category | 5 | [0.1, 1000] | mm | Generally does not need to be changed. It should be greater than the point spacing of the workpiece Point Cloud and smaller than the minimum distance between the workpiece Point Cloud and the noise Point Cloud |
| Minimum Point Count Threshold | Point Cloud clusters with fewer points than this value will be filtered out | 100 | [1,10000000] | / | Generally does not need to be changed. Increase the Minimum Point Count Threshold according to the amount of noise in the instance Point Cloud |
| Maximum Point Count Threshold | Point Cloud clusters with more points than this value will be filtered out | 100000 | [1,10000000] | / | Generally does not need to be changed. If the number of workpiece Point Cloud points exceeds 100000, increase the Maximum Point Count Threshold |
| Select Top ROI Point Cloud | If selected, calculates and sorts the average Z coordinate of the same category of Point Cloud under the ROI coordinate system, and retains the Point Cloud category with the largest average Z coordinate (top Point Cloud). If not selected, retains all Point Clouds that meet the conditions | Not selected | / | / | If the workpiece Point Cloud is above the noise Point Cloud, selecting this option retains the workpiece Point Cloud; if the workpiece Point Cloud is below the noise Point Cloud, you should also adjust the Z-axis of the ROI coordinate system to point downward when selecting this option so that the workpiece Point Cloud can be retained |
| Visualize Process Data | If selected, saves the denoised Point Cloud, which can be found in C:_data | Not selected | / | / | In debugging mode, select this option if you need to save visualization data |
- Example
2.1.2 Point Cloud Downsampling
- Function
Samples the Point Cloud according to the specified point spacing during downsampling
- Applicable scenario
If excessive Camera precision causes the instance Point Cloud quantity to be too large and the log reports an error, this option should be selected
- Parameter description
| Parameter | Description | Default Value | Parameter Range | Unit |
|---|---|---|---|---|
| Downsampling Point Spacing (mm) | Samples the Point Cloud according to the specified point spacing | 5.0 | [0.1, 1000] | mm |
Parameter tuning
- Set it according to the point spacing of the instance Point Cloud. The larger the value, the fewer Point Cloud points remain after downsampling
2.1.3 Calculate Normals
- Function
Calculates Point Cloud normals for subsequent Point Cloud processing
- Parameter description
| Parameter Name | Description | Default Value | Range |
|---|---|---|---|
| Fix Normal Orientation | Whether to fix the orientation when calculating normals. After enabling, the normal orientation is determined by the orientation reference vector | Selected | / |
| Neighborhood Point Count for Normal Calculation | The larger the value, the more neighboring points are referenced, but local changes may be ignored; the smaller the value, the opposite applies | 30 | [1,200] |
| Orientation Reference Vector | Orientation reference vector for normal calculation | [0,0,1] |
- Parameter tuning
Cannot be changed
2.1.4 Point Cloud Contour Extraction
- Function
Extracts the workpiece contour from the instance Point Cloud
- Applicable scenario
When using 2.2.4 **Enable Contour Mode **, Point Cloud Contour Extraction should also be selected
- Parameter description
| Parameter Name | Description | Default Value | Range | Unit | Tuning Suggestion |
|---|---|---|---|---|---|
| Reference Radius (mm) | The search radius for extracting contours from the instance Point Cloud | 10 | [0.1,10000000000] | mm | The reference radius is recommended to be set to 1/2 of the downsampling point spacing in 2.1.2Point Cloud Downsampling, and it must be greater than the Point Cloud spacing |
| Point Cloud Contour Search Mode | Normal Mode | Normal Mode; Plane Mode |
- Example
2.1.5 Filter Point Cloud by hsv Color Range (Hue, Saturation, Value)
- Function
Filters the Point Cloud by hue, saturation, and brightness in the Point Cloud image to screen out Point Cloud regions that match the target range
- Parameter description
| Parameter Name | Description | Default Value | Range |
|---|---|---|---|
| Filter Depth by hsv - Maximum Color Range Value | Maximum color value for filtering the Point Cloud | [0.9,0.9,0.9] | [[0,0,0],[1,1,1]] |
| Filter Depth by hsv - Minimum Color Range Value | Minimum color value for filtering the Point Cloud | [0.0,0.0,0.0] | [[0,0,0],[1,1,1]] |
- Example
2.1.6 Filter Point Cloud by Three-channel Color
- Function
Filters the Point Cloud by three-channel color to screen out Point Cloud regions that match the target range
- Parameter description
| Parameter Name | Description | Default Value | Range |
|---|---|---|---|
| Filter Point Cloud by Three-channel Color - Maximum Color Value | Maximum color value for filtering the Point Cloud | [0.9,0.9,0.9] | [[0,0,0],[1,1,1]] |
| Filter Depth by Three-channel Color - Minimum Color Value | Minimum color value for filtering the Point Cloud | [0.0,0.0,0.0] | [[0,0,0],[1,1,1]] |
- Example
2.1.7 Select Point Cloud Within the ROI Region
- Function
Selects the Point Cloud within the ROI 3D region from the instance Point Cloud. This default function cannot be deleted
- Example
2.1.8 Remove Points Whose Normal Exceeds the Angle Threshold
- Function
Removes Point Cloud points whose angle between the normal vector and the axis direction of the standard normal vector is greater than the normal vector angle threshold
- Applicable scenario
Based on ordered quadrilateral loading and unloading
- Parameter description
| Parameter Name | Description | Default Value | Range | Unit |
|---|---|---|---|---|
| Angle Threshold | Point Cloud points greater than this angle threshold are considered to belong to different instances | 15 | [-360, 360] | / |
| Standard Normal Axis Direction | The angle formed between the Point Cloud normal vector and the axis direction of the standard normal vector | Z-axis | X/Y/Z-axis | / |
| Whether to Use the ROI Coordinate System | If selected, calculates the angle between the normal vector and the axes of the ROI coordinate system; if not selected, calculates the angle between the normal vector and the axes of the Camera coordinate system | Not selected | / | / |
- Parameter tuning
2.1.9 Point Cloud Plane Segmentation
- Function
Retains or removes the plane with the largest number of Point Cloud points in the instance Point Cloud
- Applicable scenario
The instance Point Cloud contains noisy planes
- Parameter description
| Parameter | Description | Default Value | Range | Unit | Tuning Suggestion |
|---|---|---|---|---|---|
| Reference Distance for Plane Fitting (mm) | If the distance between a point and the plane is lower than the reference distance, the point is considered to be on the plane; otherwise, it is considered to be outside the plane | 3 | [0.001，10000] | mm | Generally unchanged |
| Remove Plane | If selected, removes the plane with the largest number of Point Cloud points; if not selected, retains the plane with the largest number of Point Cloud points | Not selected | / | / | If the plane with the largest number of Point Cloud points is the workpiece, retain the plane and do not select this option; if the plane with the largest number of Point Cloud points is noise, remove the plane and this option should be selected |
- Example
2.1.9 Point Cloud Outlier Removal
- Function
Identifies and removes outlier noise from the Point Cloud to improve Point Cloud quality
- Applicable scenario
The instance Point Cloud contains many outlier noise points
- Parameter description
| Parameter Name | Description | Default Value | Range |
|---|---|---|---|
| Reference Neighborhood Point Count | The number of neighboring points around each point in the Point Cloud, that is, the neighborhood size. For dense Point Clouds, even a small neighborhood is sufficient to reflect workpiece features, so a smaller value can be used; for sparser Point Clouds, a larger neighborhood is required to reflect workpiece features, so a larger value should be used. | 30 | [1, 10000000] |
| Standard Deviation Multiplier | Used to identify outlier noise. If the deviation between a point coordinate and the average coordinate value of the instance Point Cloud exceeds the standard deviation multiplier, the point is considered an outlier. The smaller the value, the more points are considered outliers and removed, but this may lead to misjudgment and removal of important workpiece features; the larger the value, the fewer points are considered outliers and removed, but some outliers may be retained and affect workpiece recognition accuracy. | 0.005 | [0.0001, 2] |
- Parameter tuning
Generally unchanged. If the Point Cloud becomes too sparse after Point Cloud Outlier Removal, increase the Standard Deviation Multiplier
- Example
2.1.10 Filter Out Point Clouds Whose Object Distance Exceeds the Limit
- Function
Filters out Point Cloud data in the specified direction, removes noise points, and improves image recognition accuracy
- Parameter description
| Parameter | Description | Default Value | Parameter Range | Unit | Tuning Suggestion |
|---|---|---|---|---|---|
| Specified Axis | The specified axis of the Point Cloud, used to filter Point Cloud data in the specified direction | Z-axis | X/Y/Z-axis | / | Specified Axis generally does not need to be changed |
| Threshold (mm) | Along the specified axis direction, if the distance between the lower-layer Point Cloud and the workpiece Point Cloud is greater than this threshold, the lower-layer Point Cloud will be filtered out; if the distance between the lower-layer Point Cloud and the workpiece Point Cloud is smaller than this threshold, the lower-layer Point Cloud will be retained | 750 | [0, 1000] | mm | Adjust the Threshold according to the actual scenario. The larger the Threshold, the fewer Point Cloud points are filtered out; the smaller the Threshold, the more Point Cloud points are filtered out |
| Select Coordinate System | Filters Point Cloud data under the selected coordinate system | ROI Coordinate System | Camera Coordinate System; ROI Coordinate System; Object Self Coordinate System | / |
- Example
2.1.11 Optimize Mask Based on Point Cloud
- Function
Based on the Point Cloud within ROI 3D, removes Point Cloud data in the Mask that is not within ROI 3D, improving the precision of the Mask
2.2 Quadrilateral Pose Estimation
2.2.1 Fitting Reference Distance (mm)
- Function
The model calculates an ideal quadrilateral based on the instance Point Cloud. Point Cloud points whose distance to the ideal quadrilateral is smaller than the fitting reference distance are fitted into a quadrilateral
- Applicable scenario
Suitable for ordered quadrilateral loading and unloading scenarios
- Parameter description
Default Value: 2
Range: [0.1, 1000]
Unit: m
Parameter tuning
- The logs for fitting a quadrilateral are shown below
If the log reports "unable to fit a quadrilateral", you should first enter the quadrilateral length and width on the workpiece interface before running again
If the log still reports "unable to fit a quadrilateral" after adding length and width priors, increase Fitting Reference Distance so that more Point Cloud points are included in the quadrilateral fitting. If the value is too large, noise points may also be included
2.2.2 Fitting Score Threshold
- Function
Calculates the ratio of the number of Point Cloud points in the fitted quadrilateral to the number of Point Cloud points in the ideal quadrilateral. Fitted quadrilaterals whose ratio is less than the fitting score threshold will be filtered out
- Applicable scenario
Quadrilateral-based ordered loading and unloading, quadrilateral-based random picking
- Parameter description
Default Value: 0.5
Range: [0,1]
- Parameter tuning
If the log reports "Quadrilateral Point Cloud detected, filtered by score" as shown below, you should reduce the Fitting Score Threshold according to the fitting score in the log
- Example
When the fitting score threshold is 0.6, instance 5 does not fit a quadrilateral, and 5 Pick Points are generated
When the fitting score threshold is 0.5, instance 5 fits a quadrilateral, and 6 Pick Points are generated
2.2.3 Plane Fluctuation Range
- Function
Allows the degree of unevenness of the main plane of the quadrilateral
- Applicable scenario
Ordered quadrilateral loading and unloading
- Parameter description
Default Value: 0.005
Range: [0, 1]
- Parameter tuning
First run with the default value. If many obviously correct objects are not detected, try increasing this value appropriately. If many obviously distorted or incorrect quadrilaterals are mistakenly recognized as targets, try decreasing this value appropriately.
Do not adjust by too much each time. For example, fine-tune in increments of 0.001 or 0.002, and then observe the change in effect.
It works together with "Fitting Score Threshold". If the score threshold is already high, the plane fluctuation range can be relaxed appropriately; if the score threshold is low, a stricter plane fluctuation range may be needed to ensure quality.
2.2.4 Object Pose Correction
Fine Matching Search Radius (mm)
- Function
During fine matching, the template Point Cloud is matched with the instance Point Cloud, and each point in the template Point Cloud needs to search for the nearest neighbor point in the instance Point Cloud. The fine matching search radius represents both the search radius in the instance Point Cloud and the distance threshold between each point in the template Point Cloud and its nearest neighbor point in the instance Point Cloud. If the distance between a point and its nearest neighbor point is smaller than the fine matching search radius, the two points are considered matchable; otherwise, they are considered unmatchable.
- Applicable scenario
Ordered loading and unloading of planar workpieces, random picking of planar workpieces, and positioning/assembly scenarios for planar workpieces
- Parameter description
Default Value: 10
Range: [1, 500]
Unit: mm
- Parameter tuning
Usually unchanged
Fine Matching Search Mode
- Function
The way the template Point Cloud retrieves the nearest neighbor point in the instance Point Cloud during fine matching
- Applicable scenario
If the fine matching effect between the template Point Cloud and the instance Point Cloud is poor, this function should be adjusted
- Parameter description
| Parameter | Description |
|---|---|
| Point-to-point | Each point in the template Point Cloud searches for its nearest neighbor point in the instance Point Cloud (the point with the shortest straight-line distance within the search radius), which is suitable for all workpieces |
| Point-to-plane | Each point in the template Point Cloud searches for its nearest neighbor point in the instance Point Cloud along its normal vector, which is suitable for workpieces with obvious geometric features |
| Combination of point-to-point and point-to-plane | First uses point-to-point mode to optimize the workpiece pose in the instance Point Cloud, and then uses point-to-plane mode to optimize the workpiece pose in the instance Point Cloud. It is suitable for workpieces with obvious geometric features
|
Use Contour Mode
- Function
Extracts contour Point Clouds from the template Point Cloud and the instance Point Cloud for coarse matching
- Applicable scenario
Ordered loading and unloading of planar workpieces, random picking of planar workpieces, and positioning/assembly scenarios for planar workpieces. If the result of coarse matching using keypoints is poor, this function should be selected so that contour Point Cloud is used again for coarse matching
- Parameter tuning
The result of coarse matching affects the result of fine matching. If the fine matching result is poor, select Use Contour Mode
Contour Search Range (mm)
- Function
The search radius for extracting contour Point Cloud data from the template Point Cloud and the instance Point Cloud
- Applicable scenario
Ordered loading and unloading of general workpieces, random picking of general workpieces, and positioning/assembly scenarios for general workpieces
- Parameter description
Default Value: 5
Range: [0.1, 500]
Unit: mm
- Parameter tuning
If the value is small, the radius for searching contour Point Cloud is small, making it suitable for extracting detailed workpiece contours, but the extracted contour may contain outlier noise;
If the value is large, the radius for searching contour Point Cloud is large, making it suitable for extracting broader workpiece contours, but the extracted contour may ignore some detailed features.
Save Pose Estimation [Fine Matching] Data
- Function
If selected, fine matching data is saved
- Applicable scenario
Ordered loading and unloading of planar workpieces, random picking of planar workpieces, positioning/assembly of planar workpieces, and positioning/assembly of planar workpieces (matching only)
- Example
Fine matching data is saved in the Project save path \ProjectFolder\data\PickLight\HistoricalDataTimestamp\Builder\pose\output folder.
2.3 Empty ROI Determination
- Function
Determines whether there are any workpieces (Point Cloud data) remaining in ROI 3D. If the number of 3D points in ROI 3D is smaller than this value, it indicates that no workpiece Point Cloud remains, and no Point Cloud is returned in this case
- Parameter description
Default Value: 1000
Range: [0, 100000]
- Usage process
Set the minimum Point Cloud count threshold for ROI 3D. If it is smaller than this threshold, the workpiece Point Cloud in ROI 3D is insufficient, and it is therefore determined that there is no workpiece in ROI 3D;
In the robot configuration, add a new vision status code to facilitate subsequent robot signal processing.