Vision-based electric connector detection system and method

Abstract

The invention discloses an electric connector detection system and method based on vision, and belongs to the field of machine vision and industrial detection, and the system comprises a large constant water star series industrial camera, a multi-angle and high-uniformity adjustable light source system, a product feeding moving and packaging mechanism, a computer control system and UI software. Wherein the computer control system adopts multi-thread scheduling, a modular architecture and a sub-pixel-level algorithm, and supports high-throughput processing of 3-6 4024 * 3036 pixel industrial cameras, an image acquisition module of the computer control system can uniformly manage multiple models of cameras, and a UI (User Interface) is a drag-and-drop interface and integrates debugging tools such as OpenGL (Open Graphics Library) and parameter hot loading; the method comprises the steps of hardware configuration, trigger starting, image acquisition, positioning correction, ROI drawing, detection, reference line determination, size measurement and sorting control, ROI drawing can adapt to regular and irregularly arranged pins, the adaptability is high, the precision is high, operation is convenient and fast, full-process automatic detection can be achieved, and the requirement of a high-speed production line is met.

Description

Technical Field

[0001] This invention relates to the fields of machine vision and industrial inspection technology, and in particular to a vision-based electrical connector inspection system and method. Background Technology

[0002] Electrical connectors, as core functional components for the transmission of electrical / optical signals between devices, components, equipment, and systems, are used in multiple key industries such as transportation, communications, automotive, and military aerospace, with particularly strong demand in automotive electronics and 5G communications. The appearance quality of connectors, such as pin pitch, height, and surface defects, directly determines their electrical performance and system stability; therefore, high-precision and standardized testing is crucial.

[0003] Currently, connector quality inspection mainly relies on manual visual inspection, but this method has significant limitations: low inspection efficiency, making it difficult to match the high-speed requirements of automated production lines; results are greatly influenced by the subjective experience of the inspectors, easily leading to missed or false detections; when dealing with the production of multiple connector models, the technical adaptability requirements for inspectors are high, resulting in high training costs; and it cannot meet the standardized inspection requirements of real-time and high precision. Under the trend of intelligent transformation, machine vision-based inspection technology is gradually being applied in this field due to its advantages such as non-contact, high speed, and high precision. However, the inspection of through-hole components (large in size and with long, dense pins) still faces multiple technical bottlenecks: visual positioning is difficult, there is a lack of stable marker points, and pin offset is difficult to calculate accurately; there are defects in light sources and image acquisition, with traditional light sources prone to reflection and overlapping and occlusion when pins are dense; the inspection efficiency and robustness are insufficient, with low recognition efficiency and weak anti-interference ability in complex scenarios; and poor adaptability to multiple models, requiring frequent parameter adjustments or hardware replacements, making it difficult to achieve "one machine for multiple uses".

[0004] Currently, some machine vision inspection technologies have been disclosed, but they all have significant limitations and cannot directly adapt to the inspection needs of electrical connectors, especially through-hole components. For example, patent CN119600032B discloses a machine vision-based industrial product quality inspection method and system. It uses an industrial camera to acquire images of electronic components, performs preprocessing and feature extraction, and then uses a deep learning model to identify defects and calculate a quality assessment index Q to determine the quality level. Although this technology achieves automated inspection of electronic components, it mainly targets defect identification of general electronic components and does not design a dedicated positioning algorithm for the characteristics of dense pins and lack of stable marker points in through-hole components of electrical connectors. The light source uses an optimized fixed layout, which cannot solve the problem of reflection interference between the connector shell and the pin area. Furthermore, the system modules are fixed configurations and lack a pluggable modular design, making it difficult to quickly adapt to the inspection needs of different models of electrical connectors.

[0005] Another patent, CN114527073A, proposes a rapid and high-precision appearance quality inspection system and method for reflective curved surfaces. It establishes a reflection model through a high-speed multi-angle light source control system, a motion control system, and a line-scan image acquisition system. The difference between specular reflection and diffuse reflection images is used to enhance defect features, thus enabling the detection of reflective curved surfaces. This technology focuses on detecting appearance defects in reflective curved surfaces and does not address the measurement of key dimensions such as pin spacing and coplanarity of electrical connectors. Its light source control scheme is optimized for curved surface reflection and cannot adapt to the feature extraction requirements of densely pinned electrical connector scenarios. Furthermore, the detection process is fixed, does not support user-defined detection items and processes, and has poor compatibility with multiple electrical connector models.

[0006] Patent CN111521554A discloses a portable electrical connector appearance quality inspection system, including a housing, fixture structure, image acquisition system, and display and control system. It uses an electric slide to drive an industrial camera to move in six directions, completing the acquisition of electrical connector appearance images. While this technology is designed for electrical connector inspection, the image acquisition module lacks a unified multi-camera management architecture and does not employ a multi-threaded task scheduling mechanism, making it difficult to meet the needs of high-speed production lines in terms of inspection efficiency. It also lacks a drag-and-drop visual programming interface, making it impossible to flexibly customize the inspection process. Furthermore, it lacks sub-pixel-level detection algorithms, resulting in limited measurement accuracy. The light source only has basic supplementary lighting functions, failing to address the reflection interference between pins and the housing, as well as the occlusion problems caused by dense pins, making it difficult to adapt to the high-precision inspection requirements of through-hole components.

[0007] In summary, existing manual inspection methods and related machine vision inspection technologies suffer from problems such as inaccurate positioning, significant glare interference, ineffective pin obstruction handling, insufficient inspection efficiency and accuracy, poor compatibility with multiple models, and low flexibility in customizing inspection processes when inspecting electrical connectors, especially through-hole components. These issues cannot meet the high precision, high speed, and strong compatibility requirements of automated production lines for electrical connector inspection. There is an urgent need for a targeted inspection system and method to address these pain points. Summary of the Invention

[0008] The purpose of this invention is to address the shortcomings of the prior art by providing a vision-based electrical connector inspection system and method, which has the advantages of high-precision positioning and measurement, rapid adaptation to multiple models, flexible customization of the inspection process, strong anti-interference ability, and high inspection efficiency.

[0009] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0010] A vision-based electrical connector inspection system includes an industrial camera, an adjustable light source system, a product loading, moving, and packaging mechanism, a computer control system, and user interface (UI) software. The industrial camera acquires image information of the electrical connectors. The adjustable light source system has a multi-angle, highly uniform structure. The product loading, moving, and packaging mechanism is used for positioning, conveying, and sorting and packaging the electrical connectors after inspection. The computer control system is the core control unit, used for image processing, algorithm calculation, result analysis, and communication with the UI software. The computer control system employs a multi-threaded task scheduling mechanism to form a millisecond-level image processing pipeline, and its image acquisition module is divided into a driver layer, a management layer, and an application layer. The UI software is a drag-and-drop visual programming interface used for customizing inspection projects, configuring parameters, displaying results, and managing data. The UI software integrates an OpenGL real-time rendering engine and has debugging tools. The UI is a user-interactive interface.

[0011] Furthermore, the industrial camera is a Daheng Mercury series industrial camera, and the computer control system supports high-throughput data processing of 3-6 4024*3036 pixel industrial cameras, and supports simultaneous image acquisition by different models of industrial cameras.

[0012] Furthermore, the computer control system adopts a modular architecture and includes 14 pluggable functional components. These pluggable functional components include image acquisition, image display, ROI drawing, pin lookup, fixed baseline, template matching, dynamic baseline, pin coplanarity measurement, corresponding pin spacing measurement, pin spacing measurement, jump statement, end statement, general I / O, and delay. The detection scheme can be quickly switched through an external description file. The ROI is a specific image area of ​​the pin to be detected during the visual inspection of the electrical connector.

[0013] Furthermore, the image acquisition module of the computer control system uses macro definitions, Maps, and a multi-camera management class with a singleton pattern to manage various types of industrial cameras in a unified manner, thereby achieving a standardized design of the main program logic.

[0014] Furthermore, the debugging tools for the UI software include parameter hot loading, process backtracking, and error heatmap tools, and the computer control system adopts a sub-pixel level detection algorithm.

[0015] A vision-based method for detecting electrical connectors includes the following steps:

[0016] Step S1: Configure the hardware through the UI. The hardware configuration includes specifying the number of industrial cameras, setting the imaging parameters of the industrial cameras and the PLC hardware address. The imaging parameters include exposure time, gain and triggering mode.

[0017] Step S2: Drag and drop to select a general-purpose I / O module, and select one I / O port of the general-purpose I / O module as the detection trigger signal to start the detection process;

[0018] Step S3: Drag and select the image acquisition algorithm module, set the image source to the industrial camera configured in step S1, and acquire the image of the electrical connector from the industrial camera;

[0019] Step S4: Use the template matching algorithm module to accurately locate the electrical connector and correct the offset of the subsequent ROI; the ROI is the specific image area of ​​the pin to be detected during the visual inspection of the electrical connector.

[0020] Step S5: Drag and select the ROI drawing algorithm module to draw the pin detection area on the image obtained in step S3. The pin detection area is slightly larger than the size of the electrical connector pin.

[0021] Step S6: Drag and select the pin detection plugin to detect the pins in the ROI area drawn in step S5. The detection includes image segmentation based on image grayscale values, area filtering, morphological processing, setting the starting position of pin detection, and viewing intermediate images of image processing.

[0022] Step S7: Drag and select the linear plug to determine the baseline for measuring the electrical connector;

[0023] Step S8: Drag and select the size measurement algorithm module. The size measurement algorithm module includes pin coplanarity measurement, pin spacing measurement and corresponding pin distance measurement modules. The pin gripping point can be manually selected and measurement constraints can be enabled to realize the size measurement of the electrical connector.

[0024] Step S9: Drag and select the image display plugin to display the size measurement result image and output OK or NG judgment result; OK means Okay, good product; NG means No Good, defective product.

[0025] Step S10: Drag and drop to select a general I / O module, select one I / O port of the output terminal of the general I / O module, and send the OK or NG judgment result signal to the PLC so that the PLC controls the product sorting.

[0026] Furthermore, the ROI drawing algorithm module described in step S5 supports two ROI generation methods: an array generation method suitable for regularly arranged pins and a template matching generation method suitable for irregularly arranged pins.

[0027] Furthermore, the morphological processing described in step S6 is achieved by setting a convolution kernel for morphological transformation, which is used to truncate the adhesive portion between the pin and the colloid and enhance the pin features. The morphological processing methods include, but are not limited to, dilation, erosion, opening operation, closing operation, top cap, and black cap. The convolution kernel is circular or square, and the area filtering is based on the number of pixels occupied by the detected pin.

[0028] Furthermore, the line-finding plugin described in step S7 provides three ways to define a baseline: a fixed baseline defining the start and end points, a dynamic baseline fitted by calipers, and a dynamic two-point line defined by capturing the feature points to obtain two dynamic feature points on the line through small-area line fitting.

[0029] Furthermore, the positions of the pin gripping points in step S8 include, but are not limited to, top, bottom, left, right, upper left, lower left, upper right, lower right, and center point, and the measurement constraints include adjacent pin height constraints and maximum pin height difference constraints.

[0030] Compared with the prior art, the present invention, employing the above technical solution, has the following beneficial effects:

[0031] (1) The vision-based electrical connector inspection system and method proposed in this invention have strong adaptability and high inspection throughput, and can flexibly meet the inspection needs of different scenarios: The system adopts a modular architecture design and integrates 14 pluggable functional components. The inspection scheme can be quickly switched through external description files; The inspection method supports array-type ROI generation of regularly arranged pins and template matching type ROI generation of irregularly arranged pins. With the help of three baseline definition methods, it can adapt to electrical connectors of different models and different pin arrangement forms; At the same time, the unified management of multiple models of industrial cameras is realized through macro definition, Map and singleton pattern multi-camera management class, which supports 3-6 4024*3036 pixel industrial cameras to acquire images and process high throughput data at the same time, meeting the production line needs of multi-view and large-batch inspection.

[0032] (2) The present invention proposes a vision-based electrical connector inspection system and method with high detection accuracy and reliable results, effectively reducing the risk of false detection and missed detection: The system adopts a sub-pixel level detection algorithm, combined with template matching for precise positioning and correction of ROI offset, to ensure the accuracy of the detection area; During the detection process, the system can cut off the adhesive part between the pin and the adhesive, enhance the pin features and filter out false detection results through operations such as image segmentation based on gray value, pixel number area screening and morphological convolution kernel processing; When measuring the size, it supports the selection of 9 pin gripping point positions, and can enable adjacent pin height constraints and maximum pin height difference constraints, further improving the accuracy of pin coplanarity, spacing and corresponding distance measurement, and ensuring the consistency and reliability of the detection results.

[0033] (3) The vision-based electrical connector inspection system and method proposed in this invention are easy to operate and efficient to debug, significantly reducing the cost of use and maintenance: The UI software adopts a drag-and-drop visual programming interface, which allows users to complete hardware configuration, inspection process construction and parameter adjustment without writing code, greatly reducing the operation threshold; At the same time, it integrates debugging tools such as parameter hot loading, process backtracking and error heat map, and supports step-by-step viewing of intermediate images of image processing, which makes it easy for users to quickly optimize inspection parameters and troubleshoot problems; The system as a whole adopts a multi-threaded task scheduling mechanism to form a millisecond-level image processing pipeline, which is combined with PLC automatic control of product sorting to realize full-process automation from image acquisition to result output, reducing manual intervention and improving inspection efficiency and production line adaptability. Attached Figure Description

[0034] Figure 1 Provide a system overall design block diagram;

[0035] Figure 2 This is a flowchart of the electrical connector testing process;

[0036] Figure 3 The image shows the user interface of the testing software.

[0037] Figure 4 This is a screenshot of the detection results from the detection software. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] In one embodiment, combined Figure 1 This invention provides a vision-based electrical connector inspection system, including an industrial camera, an adjustable light source system, a product loading, moving, and packaging mechanism, a computer control system, and UI software. The industrial camera is used to acquire image information of the electrical connectors. The adjustable light source system has a multi-angle, highly uniform structure. The product loading, moving, and packaging mechanism is used for positioning, conveying, and sorting and packaging the electrical connectors after inspection. The computer control system is the core control unit, used for image processing, algorithm calculation, result analysis, and communication with the UI software. The computer control system adopts a multi-threaded task scheduling mechanism to form a millisecond-level image processing pipeline, and its image acquisition module is divided into a driver layer, a management layer, and an application layer. The UI software is a drag-and-drop visual programming interface used for customizing inspection projects, configuring parameters, displaying results, and managing data. The UI software integrates an OpenGL real-time rendering engine and has debugging tools (such as...). Figure 3 As shown, the interface includes a project workflow setup area, parameter configuration area, operation log area, and functional module library, intuitively presenting the implementation of drag-and-drop operation; the UI is the user interaction interface.

[0040] Furthermore, the industrial camera is a Daheng Mercury series industrial camera, and the computer control system supports high-throughput data processing of 3-6 4024*3036 pixel industrial cameras, and supports simultaneous image acquisition by different models of industrial cameras.

[0041] Furthermore, the computer control system adopts a modular architecture and includes 14 pluggable functional components. These pluggable functional components include image acquisition, image display, ROI drawing, pin lookup, fixed baseline, template matching, dynamic baseline, pin coplanarity measurement, corresponding pin spacing measurement, pin spacing measurement, jump statement, end statement, general I / O, and delay. The detection scheme can be quickly switched through an external description file. The ROI is a specific image area of ​​the pin to be detected during the visual inspection of the electrical connector.

[0042] Furthermore, the image acquisition module of the computer control system manages various types of industrial cameras in a unified manner through macro definitions, Maps, and a multi-camera management class with a singleton pattern, thereby achieving a standardized design of the main program logic.

[0043] Furthermore, the debugging tools of the UI software include parameter hot loading, process backtracking, and error heatmap tools, and the computer control system adopts a sub-pixel level detection algorithm to improve detection accuracy.

[0044] In one embodiment, combined Figure 2 A vision-based method for detecting electrical connectors is provided, the method comprising the following steps:

[0045] Step S1: Configure the hardware through the UI. The hardware configuration includes specifying the number of industrial cameras, setting the imaging parameters of the industrial cameras and the PLC hardware address. The imaging parameters include exposure time, gain and triggering mode.

[0046] Step S2: Drag and drop to select a general-purpose I / O module, and select one I / O port of the general-purpose I / O module as the detection trigger signal to start the detection process;

[0047] Step S3: Drag and select the image acquisition algorithm module, set the image source to the industrial camera configured in step S1, and acquire the image of the electrical connector from the industrial camera;

[0048] Step S4: The template matching algorithm module is used to accurately locate the electrical connector and correct the offset of the subsequent ROI to ensure the accuracy of the detection area; the ROI is the specific image area of ​​the pin to be detected during the visual inspection of the electrical connector.

[0049] Step S5: Drag and select the ROI drawing algorithm module to draw the pin detection area on the image obtained in step S3. The pin detection area is slightly larger than the size of the electrical connector pin.

[0050] Step S6: Drag and select the pin detection plugin to detect the pins in the ROI area drawn in step S5. The detection includes image segmentation based on image grayscale values, area filtering, morphological processing, setting the starting position of pin detection, and viewing intermediate images of image processing.

[0051] Step S7: Drag and select the linear plug to determine the baseline for measuring the electrical connector;

[0052] Step S8: Drag and select the size measurement algorithm module. The size measurement algorithm module includes pin coplanarity measurement, pin spacing measurement and corresponding pin distance measurement modules. The pin gripping point can be manually selected and measurement constraints can be enabled to realize the size measurement of the electrical connector.

[0053] Step S9: Drag and select the image display plugin to display the size measurement result image and output OK or NG judgment result (as shown in Figure 4, which clearly presents key information such as pin size data, coplanarity judgment result, system operation log and parameter deviation range).

[0054] Step S10: Drag and drop to select a general I / O module, select one I / O port of the general I / O module output terminal, and send the OK or NG judgment result signal to the PLC. The PLC controls the product sorting to realize automatic sorting after detection.

[0055] Furthermore, the ROI drawing algorithm module described in step S5 supports two ROI generation methods: an array generation method suitable for regularly arranged pins and a template matching generation method suitable for irregularly arranged pins, which can adapt to electrical connectors with different pin arrangement forms.

[0056] Furthermore, the morphological processing in step S6 is achieved by setting a convolution kernel for morphological transformation, which is used to truncate the adhesive portion between the pin and the colloid and enhance the pin features. The morphological processing methods include, but are not limited to, dilation, erosion, opening operation, closing operation, top cap, and black cap. The convolution kernel is circular or square. The area filtering is based on the number of pixels occupied by the detected pin to reduce false detections.

[0057] Furthermore, the line-finding plugin described in step S7 provides three ways to define a baseline: a fixed baseline defining the start and end points, a dynamic baseline fitted by calipers, and a dynamic two-point line defined by dragging a threshold to capture feature points, providing a precise baseline for dimension measurement.

[0058] Furthermore, the positions of the pin gripping points in step S8 include, but are not limited to, top, bottom, left, right, upper left, lower left, upper right, lower right, and center point. The measurement constraints include adjacent pin height constraints and maximum pin height difference constraints to ensure the accuracy of dimensional measurements.

[0059] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A vision-based electrical connector detection system, characterized by, The application relates to an industrial camera, an adjustable light source system, a product loading, moving and packaging mechanism, a computer control system and UI software; the industrial camera is used for acquiring image information of an electric connector; the adjustable light source system is a multi-angle and high-uniformity structure; the product loading, moving and packaging mechanism is used for positioning, conveying and sorting and packaging of the electric connector after detection; the computer control system is a core control unit and is used for image processing, algorithm operation, result analysis and communication with the UI software; the computer control system adopts a multi-thread task scheduling mechanism and constitutes a millisecond-level image processing pipeline, and an image acquisition module of the computer control system is divided into a driving layer, a management layer and an application layer; the UI software is a drag-type visual programming interface and is used for detection item customization, parameter configuration, result display and data management; the UI software integrates an OpenGL real-time rendering engine and has a debugging tool; and the UI is a user interaction interface.

2. A vision-based electrical connector detection system according to claim 1, wherein, The industrial camera is a Dahengshuiwei series industrial camera, and the computer control system supports high-throughput data processing of 3-6 4024*3036 pixel industrial cameras and supports simultaneous image acquisition of different models of industrial cameras.

3. A vision-based electrical connector detection system according to claim 1, wherein, The computer control system adopts a modular architecture and contains 14 pluggable functional components, the pluggable functional components include image acquisition, image display, ROI drawing, pin searching, fixed reference line, template matching, dynamic reference line, pin coplanarity measurement, corresponding pin spacing measurement, pin spacing measurement, jump statement, end statement, general I / O and time delay, and rapid switching of detection schemes is realized through an external description file; the ROI is a specific image area of a pin to be detected in the electric connector visual detection process.

4. The vision-based electrical connector inspection system of claim 1, wherein, The image acquisition module of the computer control system is uniformly managed for various models of industrial cameras through a multi-camera management class of macro definition, Map and singleton mode, and regular design of main program logic is realized.

5. The vision-based electrical connector detection system of claim 1, wherein, The debugging tool of the UI software includes parameter hot loading, process backtracking and error heat map tools, and the computer control system adopts a sub-pixel level detection algorithm.

6. A vision-based electrical connector detection method, suitable for use in a vision-based electrical connector detection system according to any one of claims 1 to 5, characterized in that, The application further discloses a method for detecting an electric connector, and the method comprises the following steps: S1, configuring hardware through a UI, wherein the hardware configuration comprises specifying the number of industrial cameras, setting imaging parameters of the industrial cameras and a PLC hardware address, and the imaging parameters include exposure time, gain and trigger mode; S2, selecting a general I / O module and selecting an IO port of an input end of the general I / O module as a detection trigger signal to start a detection process; S3, selecting an image acquisition algorithm module and setting an image source as the industrial camera configured in S1 to acquire images of the electric connector from the industrial camera; S4, adopting a template matching algorithm module to accurately position the electric connector and correct the offset of a subsequent ROI; the ROI is a specific image area of a pin to be detected in the electric connector visual detection process; S5, selecting a ROI drawing algorithm module and drawing a pin detection area on the image acquired in S3, wherein the pin detection area is slightly larger than the size of the electric connector pin. Step S6, drag to select the pin detection plug-in, detect the pins in the ROI area drawn in step S5, the detection includes image segmentation based on image gray value, area screening, morphological processing, setting the starting position of pin detection and viewing image processing intermediate image; Step S7, drag to select the straight line finding plug-in to determine the reference line of the electrical connector measurement; Step S8, drag to select the size measurement algorithm module, the size measurement algorithm module includes pin coplanarity measurement, pin spacing measurement and corresponding pin distance measurement module, manually select pin grabbing points and enable measurement constraints to realize electrical connector size measurement; Step S9, drag to select the image display plug-in to display the size measurement result image and output OK or NG judgment result, OK is Okay, good product; NG is No Good, defective product; Step S10, drag to select the general I / O module, select one IO port of the general I / O module output end, send the OK or NG judgment result signal to PLC, and control product sorting by PLC.

7. The vision-based electrical connector detection method of claim 6, wherein, The drawing ROI algorithm module in step S5 supports two ROI generation methods, which are array generation method suitable for regular arrangement of pins and template matching generation method suitable for irregular arrangement of pins.

8. The vision-based electrical connector detection method of claim 6, wherein, The morphological processing in step S6 is realized by setting the convolution kernel of morphological transformation, which is used to truncate the adhesion part of the pin and the colloid and enhance the pin feature, the morphological processing method includes but is not limited to dilation, erosion, opening operation, closing operation, top hat, black hat, the convolution kernel is circular or square, and the area screening is based on the number of pixels occupied by the detected pin.

9. The vision-based electrical connector detection method of claim 6, wherein, The straight line finding plug-in in step S7 provides three kinds of reference line definition methods, which are fixed reference line defined by starting point and ending point, dynamic reference line fitted by caliper method, and dynamic two-point straight line defined by grabbing feature points through small area straight line fitting.

10. The vision-based electrical connector detection method of claim 6, wherein, The position of the pin grabbing point in step S8 includes but is not limited to upper, lower, left, right, left upper, left lower, right upper, right lower and center point, and the measurement constraint includes adjacent pin height constraint and maximum pin height difference constraint.

Images