Industrial intelligent robot
Through the image acquisition and processing module, combined with sensors and machine learning models, key feature extraction and parameter adjustment in bearing processing are realized, and the identification accuracy and stability problems in traditional technologies are solved, and the machining accuracy and automation level are improved.
Patent Information
- Application Number
- CN202411243091.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-05
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-09-05
AI Technical Summary
Traditional image recognition technology is difficult to accurately extract key features in bearing processing, resulting in unstable machining parameter setting and adjustment, lack of real-time feedback adjustment mechanism, affecting processing accuracy and product quality, and the processing process relies on manual operation and lacks intelligent control.
The image acquisition module and image adjustment processing module are adopted to collect bearing image data through industrial cameras, and the image processing workstation and sensors are used to perform preliminary image features extraction, fine parameter adjustment and intelligent optimization. Combined with machine learning models, real-time data monitoring and parameter adjustment are realized. Form closed-loop control.
It improves image recognition accuracy, realizes accurate adjustment and stability of processing parameters, realizes automated and intelligent control of bearing processing, reduces manual intervention, and improves production efficiency and product quality.
Smart Images

Figure CN119068210A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of image recognition for bearing processing, and in particular to an industrial intelligent robot. Background Art
[0002] With the transformation, upgrading and intelligent development of the manufacturing industry, industrial intelligent robots are playing an increasingly important role in the production line. As one of the key components of industrial robots, the processing quality and precision of bearings directly affect the overall performance and life of the robots. Therefore, industrial intelligent robot bearing processing technology has become a research hotspot in the manufacturing industry. In the field of industrial intelligent robot bearing processing, image recognition technology is gradually becoming a key means to improve processing accuracy and efficiency. Through high-precision image acquisition and intelligent analysis, robots can more accurately identify various features in the bearing processing process, thereby optimizing processing parameters and improving processing quality.
[0003] Traditional image recognition methods often find it difficult to accurately extract key features when processing complex and changeable bearing processing images, resulting in limited recognition accuracy, which in turn affects the setting and adjustment of processing parameters. In addition, due to the lack of an effective mapping relationship between image features and processing parameters, existing technologies often find it difficult to accurately adjust processing parameters based on image recognition results, resulting in unstable processing effects and difficulty in meeting the processing requirements of high-precision bearings. During the processing process, existing technologies often lack a real-time and effective feedback adjustment mechanism and are unable to adjust processing parameters in real time based on processing effects, resulting in accumulated processing errors and affecting the quality of the final product. In addition, most traditional processing equipment relies on manual operation and judgment, lacks intelligent control, and cannot fully utilize the advantages of image recognition technology to realize the automation and intelligence of the processing process. Summary of the invention
[0004] The purpose of the present invention is to provide an industrial intelligent robot that solves the problems raised in the above background technology.
[0005] To achieve the above object, the present invention provides the following technical solution, which is applied to image processing in bearing processing; It includes an image acquisition module and an image adjustment processing module; The specific implementation steps are as follows: Data acquisition: Through the image acquisition module, image data of the bearing processing area is collected and preliminarily extracted on the basis of no obstruction and covering the entire bearing processing area, and key features in the image data are collected and preliminarily extracted, including area ratio and edge complexity; Data adjustment processing, based on the image data acquisition and preliminary extraction of the image acquisition module, and through the image adjustment processing module, the adjusted processing parameters are transmitted to the industrial intelligent robot to perform the adjustment operation of the bearing processing direction. The image adjustment processing module includes an image feature preliminary extraction processing unit, a fine processing parameter adjustment unit and an image intelligent adjustment optimization unit; Data upload: upload image data to the industrial intelligent robot control system terminal; The equipment used in the image acquisition module includes an industrial camera for collecting image data of the bearing processing area; The equipment used in the image adjustment and processing module includes an image processing workstation and a sensor. The image processing workstation uses an image processing algorithm and a machine learning model to adjust the image. The sensor monitors real-time data during the bearing processing process, including cutting force and temperature.
[0006] Optionally, the calculation formula of the image feature preliminary extraction processing unit is as follows: ; MB=MM / ZM; Y=TY / MY; in: TC is the image characteristic value; MB is the image area ratio, MM is the number of extracted image areas, and ZM is the total number of image areas; BF is the edge complexity value, which reflects the complexity of the edge of the image within the extracted image area MM; Y is the color contrast value; TY is the color value of the extracted image, which reflects the degree of image color presentation within the extracted image area MM; MY is the image color value, and MY reflects the overall image background presentation degree within the total image area ZM; L is the image brightness value; The extraction process of the image feature preliminary extraction processing unit is as follows: Through the image acquisition module, key features including image area ratio MB, edge complexity value BF, color contrast value Y and image brightness value L are collected and preliminarily extracted, and after being input into the image feature preliminary extraction processing unit for processing, an image feature value TC for evaluating the significance and quality of the bearing processing area in the image data is output.
[0007] Optionally, the calculation formula of the fine processing parameter adjustment unit is as follows: ; in: TZ is the adjusted fine parameter value, including cutting speed and feed rate; BJ is the standard processing parameter value; a is a natural value, 1+a -TC*Y It reflects the change in sensitivity of the fine parameter value TZ after adjustment when the image characteristic value TC changes; The adjustment process of the fine processing parameter adjustment unit is as follows: Based on the image feature value TC output by the image feature preliminary extraction processing unit, combined with the standard processing parameter value BJ, and the sensitivity change of the adjusted fine parameter value TZ when the image feature value TC changes, the adjusted fine parameter value TZ adapted to the current image feature is output.
[0008] Optionally, the calculation formula of the image intelligent adjustment optimization unit is as follows: ; in: MB new is the ratio of image area after adjustment; b is the adjustment factor; BTZ is the target fine parameter value; It is used to feedback and adjust the image area ratio MB; The adjustment and optimization process of the image intelligent adjustment and optimization unit is as follows: Based on the image feature value TC output by the image feature preliminary extraction processing unit and the adjusted fine parameter value TZ output by the fine processing parameter adjustment unit, and combined with the target fine parameter value BTZ, the adjustment factor b is adjusted to output the adjusted image area ratio MB that affects the next intelligent robot processing direction recognition. new .
[0009] Optionally, based on the adjusted image area ratio MB output by the image intelligent adjustment optimization unit new The recognition adjustment is as follows: The adjusted image area ratio MB output by the image intelligent adjustment optimization unit at any time new , are used as the input value in the next image feature preliminary extraction processing unit, that is, the image area ratio MB. In other words, the adjusted image area ratio MB outputted this time new , replacing the image area ratio MB input next time, and performing iterative processing, so as to continuously identify the adjusted image area ratio MB, so as to enable the industrial robot to identify and process the bearing processing area direction.
[0010] Optionally, the adjusted image area ratio MB new The control and response of the adjustment range are as follows: S1. Set MB maxis the maximum ratio of image area, and sets MB min is the minimum ratio of image area; S2, if the image area maximum ratio MB max Greater than the adjusted image area ratio MB new Greater than the minimum image area ratio MB min , then it is considered that the adjusted image area ratio MB new It is within the safe adjustment range and should be maintained and adjusted continuously. S3, if the adjusted image area ratio MB new Greater than the maximum image area ratio MB max , if the adjusted image area ratio MB new Smaller than the minimum image area ratio MB min Regardless of the above situation, the adjusted image area ratio MB new When in a dangerous adjustment range, the sensor monitors and issues a warning, stops adjustment, and takes action.
[0011] Optionally, the edge complexity value BF specifically adopts edge detection algorithms, including Canny and SOBEL to identify edges in image data, and detects edges by calculating the gradient strength and direction of pixels in the image data and their neighborhood pixels, and extracts features from the detected edges. The features include the number, length, curvature, and direction of edges. The features are used to quantify the complexity of the edges, and the edge complexity value BF is output using algorithms including weighted summation and principal component analysis.
[0012] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention introduces advanced image recognition algorithms and deep learning technology, and combines it with an image feature preliminary extraction processing unit to more accurately extract key features in bearing processing images, such as area ratio, edge complexity, color contrast and brightness level, providing a reliable basis for setting and adjusting processing parameters, thereby improving image recognition accuracy.
[0013] 2. The present invention is based on the calculation of the fine processing parameter adjustment unit, combines the image recognition results with the preset processing parameter standards, obtains the optimal processing parameters through fine calculation, realizes the precise adjustment of the processing parameters, and improves the processing accuracy and stability.
[0014] 3. The present invention introduces a feedback adjustment mechanism through an image intelligent adjustment optimization unit, dynamically adjusts the area ratio and other key parameters according to the deviation between the real-time processing effect and the expected target, forms a closed-loop control, and ensures that the processing process always remains in the optimal state.
[0015] 4. The present invention combines image recognition technology with industrial intelligent robots to achieve automated and intelligent control of the processing process. The robot can adjust the algorithm according to the image recognition results and processing parameters, complete the processing tasks independently, reduce human intervention, improve production efficiency and product quality, and enhance the level of intelligence. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 A flow chart of the method for this industrial intelligent robot; Figure 2 A schematic diagram of the structure of the image processing module based on FIG. Figure 3 This is a schematic diagram of a device for using the image acquisition module of the present invention; Figure 4 This is a schematic diagram of equipment using the image processing module of the present invention. DETAILED DESCRIPTION
[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0018] Regarding the recognition of the processing direction of this industrial intelligent robot, it is different from the recognition of the processing direction of the existing industrial intelligent robot. The existing recognition of processing direction often finds it difficult to accurately extract key features, resulting in limited recognition accuracy, which in turn affects the setting and adjustment of processing parameters. Moreover, due to the lack of an effective mapping relationship between image features and processing parameters, the existing technology often finds it difficult to accurately adjust the processing parameters according to the image recognition results, resulting in unstable processing effects and difficulty in meeting the processing requirements of high-precision bearings. During the processing process, the existing technology often lacks a real-time and effective feedback adjustment mechanism, and is unable to adjust the processing parameters in real time according to the processing effects, resulting in the accumulation of processing errors and affecting the quality of the final product. In addition, most traditional processing equipment relies on manual operation and judgment, lacks intelligent control, and cannot fully utilize the advantages of image recognition technology to realize the automation and intelligence of the processing process. This algorithm unit is embodied in improving processing accuracy, improving processing efficiency, enhancing the level of intelligence and promoting independent development. By introducing advanced technologies, optimizing processing processes, and strengthening technology research and development and innovation measures, it can effectively solve the shortcomings of existing technologies and promote the continuous progress and development of industrial intelligent robot bearing processing technology.
[0019] For examples, see Figures 1 to 4 ,This implementation provides an industrial intelligent robot for image processing in bearing processing; It includes an image acquisition module and an image adjustment processing module; The specific implementation steps are as follows: Data acquisition: Through the image acquisition module, image data of the bearing processing area is collected and preliminarily extracted on the basis of no obstruction and covering the entire bearing processing area, and key features in the image data are collected and preliminarily extracted, including area ratio and edge complexity; Data adjustment processing, based on image data acquisition and preliminary extraction of the image acquisition module, and through the image adjustment processing module, the adjusted processing parameters are transmitted to the industrial intelligent robot to perform the adjustment operation of the bearing processing direction. The image adjustment processing module includes an image feature preliminary extraction processing unit, a fine processing parameter adjustment unit and an image intelligent adjustment optimization unit; Data upload: upload image data to the industrial intelligent robot control system terminal; The equipment used in the image acquisition module includes an industrial camera, which is used to collect image data of the bearing processing area; The equipment used in the image adjustment and processing module includes an image processing workstation and sensors. The image processing workstation uses image processing algorithms and machine learning models to adjust and process the image. The sensors monitor real-time data during the bearing processing process, including cutting force and temperature.
[0020] In this embodiment, the system combines TC, TZ and MB through the cooperation of three algorithm units in the identification of the bearing processing direction applied to the industrial intelligent robot. new The three calculation results together constitute the core part of the intelligent processing control system, realizing the intelligent, adaptive and continuous optimization of the processing process. TC is the image eigenvalue. By combining the area ratio, edge complexity, color contrast and brightness level, a comprehensive image eigenvalue is calculated to evaluate the significance and quality of the bearing processing area in the image, providing a basis for the subsequent adjustment of processing parameters. TZ is the fine parameter value after adjustment. By combining the image eigenvalue TC, the standard processing parameter value BJ and the image area ratio MB in a nonlinear way, the nonlinear effect of color contrast on the adjustment amplitude is considered to calculate the fine processing parameters that adapt to the current image characteristics to meet different processing requirements. MB new To adjust the image area ratio, by comparing the difference between the current fine processing parameters and the target parameters, the square root of the image feature value TC and is used as the adjustment weight to adjust the area ratio. This adjustment can affect the result of the next image recognition and processing, forming a closed-loop optimization system, and MB newThe calculation results can also affect the calculation of TC and TZ, and the iterative image area ratio MB makes the three algorithms of this system have a high correlation and entanglement, so that the overall algorithm system can perform automatic feedback and optimization according to actual conditions to be closer to reality.
[0021] See also Figures 1 to 4 , the calculation formula of the image feature preliminary extraction processing unit is as follows: ; MB=MM / ZM; Y=TY / MY; in: TC is the image characteristic value; MB is the image area ratio, MM is the number of extracted image areas, and ZM is the total number of image areas; BF is the edge complexity value, which reflects the complexity of the edge of the image within the extracted image area MM; Y is the color contrast value; TY is the color value of the extracted image, which reflects the degree of image color presentation within the extracted image area MM; MY is the image color value, and MY reflects the overall image background presentation degree within the total image area ZM; L is the image brightness value; The extraction process of the image feature preliminary extraction processing unit is as follows: Through the image acquisition module, key features including image area ratio MB, edge complexity value BF, color contrast value Y and image brightness value L are collected and preliminarily extracted, and after being input into the image feature preliminary extraction processing unit for processing, an image feature value TC for evaluating the significance and quality of the bearing processing area in the image data is output.
[0022] In this embodiment: first, MB in this algorithm unit is the image area ratio. Through the image processing algorithm, the extracted image area number MM and the total image area ZM are identified, and MB=MM / ZM is used to output the image area ratio MB, which is the initial image feature area. The image area ratio MB reflects the significance of the bearing processing area in the image and is an important indicator for evaluating the position and size of the processing area. Then, the edge detection algorithm is used to identify the edge of the bearing processing area, and based on the number, length, and curvature characteristics of the edge, the edge complexity value BF is calculated and output. The edge complexity value BF measures the processing area. The complexity of the regional edge helps to identify edge details and is of great significance for accurately controlling the processing path and depth. Y is the color contrast value, which is obtained by extracting the contrast difference between the image color value TY and the image color value MY. L is the image brightness value, which is obtained by calculating the average brightness value of the image, that is, the statistical information of the brightness histogram. The color contrast value Y and the image brightness value L take into account the color and brightness information of the image, which helps to distinguish the target from the background and improve the accuracy of image recognition. After the input of multiple feature image data, the image feature value TC that provides a comprehensive basis for the subsequent adjustment of processing parameters can be output; This algorithm unit enables the industrial intelligent robot to more accurately identify the bearing processing area by accurately calculating the area ratio and edge complexity, reducing the possibility of misidentification and missed identification, and improving recognition accuracy. The consideration of color contrast and brightness level enables the industrial intelligent robot to adapt to the processing environment under different lighting conditions, improve the robustness of the system, and enhance adaptability. The calculation of the preliminary image eigenvalue TC provides an important basis for the fine adjustment of subsequent processing parameters, which helps to realize the intelligent control of the processing process.
[0023] See also Figures 1 to 4 , the calculation formula of the fine processing parameter adjustment unit is as follows: ; in: TZ is the adjusted fine parameter value, including cutting speed and feed rate; BJ is the standard processing parameter value; a is a natural value, 1+a -TC*Y It reflects the change in sensitivity of the fine parameter value TZ after adjustment when the image characteristic value TC changes; The adjustment process of the fine processing parameter adjustment unit is as follows: Based on the image feature value TC output by the image feature preliminary extraction processing unit, combined with the standard processing parameter value BJ, and the sensitivity change of the adjusted fine parameter value TZ when the image feature value TC changes, the adjusted fine parameter value TZ adapted to the current image feature is output.
[0024] In this embodiment, first, TZ is the adjusted fine parameter value, which directly affects the machining quality and efficiency of the bearing. By comprehensively considering the image characteristic value TC, the standard machining parameter value BJ and the image area ratio MB, the adjusted fine parameter value TZ can more accurately adapt to the current machining conditions and improve the machining accuracy and efficiency. Among them, the natural value a is used to introduce a nonlinear relationship, so that when the image characteristic value TC increases, the adjustment of the adjusted fine parameter value TZ is more sensitive. This nonlinear relationship helps to better adapt to different image features. Specifically, the exponential function of the natural value a makes the adjustment range of the adjusted fine parameter value TZ larger when the image characteristic value TC increases, which helps to achieve more precise machining control when the image features are significant. This algorithm unit comprehensively considers multiple image features and standard parameters, and the adjustment of the fine parameter value TZ can more accurately match the actual processing requirements, thereby improving the processing accuracy and consistency. While ensuring the processing accuracy, by adjusting the processing parameters such as cutting speed and feed rate, it can optimize the processing process and improve the processing efficiency. The nonlinear relationship introduced by the exponential function of the natural value a makes the adjustment of the fine parameter value TZ more flexible and sensitive, which helps to cope with complex processing scenarios and changes and enhance the flexibility of the system.
[0025] See also Figures 1 to 4 , the calculation formula of the image intelligent adjustment optimization unit is as follows: ; in: MB new is the ratio of image area after adjustment; b is the adjustment factor; BTZ is the target fine parameter value; It is used to feedback and adjust the image area ratio MB; The adjustment and optimization process of the image intelligent adjustment and optimization unit is as follows: Based on the image feature value TC output by the image feature preliminary extraction processing unit and the adjusted fine parameter value TZ output by the fine processing parameter adjustment unit, and combined with the target fine parameter value BTZ, the adjustment factor b is adjusted to output the adjusted image area ratio MB that affects the next intelligent robot processing direction recognition. new .
[0026] In this embodiment, the algorithm unit first outputs the adjusted image area ratio MB based on the image feature value TC output by the image feature preliminary extraction processing unit and the adjusted fine parameter value TZ output by the fine processing parameter adjustment unit. new , adjusted image area ratio MB newFor the next iteration and processing cycle, it is adjusted based on the image area ratio MB, the adjustment factor b, the difference between the adjusted fine parameter value TZ and the target fine parameter value BTZ, and the square root of the image feature value TC. This adjustment method takes into account the deviation of the processing parameters and the importance of the image features, and aims to improve the subsequent image recognition and processing process by iteratively optimizing the area ratio; In the industrial intelligent robot bearing processing, the adjusted image area ratio MB new It represents the area ratio of the target area after optimization based on the current image features and processing parameters. Its significance in entering the next iteration of the image feature preliminary extraction processing unit is that by continuously adjusting the area ratio, the industrial intelligent robot can more accurately identify and process the bearing processing area, thereby improving the processing accuracy and efficiency. This loop feedback mechanism helps to achieve adaptive optimization of the processing process.
[0027] See also Figures 1 to 4 , based on the adjusted image area ratio MB output by the image intelligent adjustment optimization unit new The recognition adjustment is as follows: The adjusted image area ratio MB output by the image intelligent adjustment optimization unit at any time new , are used as the input value in the next image feature preliminary extraction processing unit, that is, the image area ratio MB. In other words, the adjusted image area ratio MB outputted this time new , replacing the image area ratio MB input next time, and performing iterative processing, thereby continuously identifying the adjusted image area ratio MB, so that the industrial robot can identify and process the bearing processing area direction; Adjusted image area ratio MB new The control and response of the adjustment range are as follows: S1. Set MB max is the maximum ratio of image area, and sets MB min is the minimum ratio of image area; S2, if the image area maximum ratio MB max Greater than the adjusted image area ratio MB new Greater than the minimum image area ratio MB min , then it is considered that the adjusted image area ratio MB new It is within the safe adjustment range and should be maintained and adjusted continuously. S3, if the adjusted image area ratio MB new Greater than the maximum image area ratio MB max , if the adjusted image area ratio MB new Smaller than the minimum image area ratio MB min Regardless of the above situation, the adjusted image area ratio MBnew When in a dangerous adjustment range, the sensor monitors and issues a warning, stops adjustment, and takes action.
[0028] In this embodiment, the algorithm unit adjusts the image area ratio MB based on the above description. new It represents the area ratio of the target area after optimization based on the current image features and processing parameters, and enters the image feature preliminary extraction processing unit for the next iteration. During the iteration process, the adjusted image area ratio MB new The output adjustment range needs to be within the safe range, which is the maximum image area ratio MB. max Greater than the adjusted image area ratio MB new Greater than the minimum image area ratio MB min If it exceeds or falls below, such as the change described in S3, the system will take corresponding measures, including sensor monitoring and warning, stopping adjustment and processing, to avoid adverse effects on the processing process. In the actual processing process, the processing effect will be evaluated regularly in real time. If the adjusted image area ratio MB new If the processing effect is not good, including the decrease of processing accuracy and the deterioration of surface quality, it is necessary to readjust the threshold range and change the adjustment strategy, and adjust the adjusted image area ratio MB. new When adjusting the image area, the stability and safety of the system should also be considered. Too frequent and large adjustments will lead to system instability and even cause safety accidents. Therefore, during the adjustment process, it is necessary to follow certain adjustment steps and frequency limits to ensure the stability and safety of the system. Specifically, sensors and control systems can be used to monitor key parameters in the processing process in real time, including cutting force and temperature, and these parameters can be fed back to the control system. The control system dynamically adjusts the adjusted image area ratio MB according to the feedback information. new , in order to achieve precise control of the machining process, this method can improve the system's adaptability and anti-interference ability.
[0029] To sum up, in the specific implementation process, by continuously iteratively adjusting the area ratio, the robot can more accurately identify and process the bearing processing area, reduce processing errors, and improve processing accuracy. The adjusted processing parameters, including cutting speed and feed rate, are more adapted to the current image features, which helps to speed up the processing speed, improve processing efficiency, and optimize processing efficiency. The feedback mechanism enables the system to cope with processing requirements under different image features, enhance the robustness and adaptability of the system, and in addition, by integrating image recognition, feature extraction, parameter adjustment, and feedback optimization steps, the industrial intelligent robot can realize the intelligence of bearing processing, improve the level of production automation and production efficiency, and achieve the effect of intelligent processing.
[0030] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. An industrial intelligent robot, characterized in that: Image processing applied to bearing processing; It includes an image acquisition module and an image adjustment processing module; The specific implementation steps are as follows: Data acquisition: Through the image acquisition module, image data of the bearing processing area is collected and preliminarily extracted on the basis of no obstruction and covering the entire bearing processing area, and key features in the image data are collected and preliminarily extracted, including area ratio and edge complexity; Data adjustment processing, based on image data acquisition and preliminary extraction of the image acquisition module, and through the image adjustment processing module, the adjusted processing parameters are transmitted to the industrial intelligent robot to perform the adjustment operation of the bearing processing direction. The image adjustment processing module includes an image feature preliminary extraction processing unit, a fine processing parameter adjustment unit and an image intelligent adjustment optimization unit; Data upload, upload the image data to the industrial intelligent robot control system terminal.
2. The industrial intelligent robot according to claim 1, characterized in that: The equipment used in the image acquisition module includes an industrial camera for collecting image data of the bearing processing area; The equipment used in the image adjustment and processing module includes an image processing workstation and a sensor. The image processing workstation uses an image processing algorithm and a machine learning model to adjust the image. The sensor monitors real-time data during the bearing processing process, including cutting force and temperature.
3. An industrial intelligent robot according to claim 2, characterized in that: The calculation formula of the image feature preliminary extraction processing unit is as follows: ; MB=MM / ZM; Y=TY / MY; in: TC is the image characteristic value; MB is the image area ratio, MM is the number of extracted image areas, and ZM is the total number of image areas; BF is the edge complexity value, which reflects the complexity of the edge of the image within the extracted image area MM; Y is the color contrast value; TY is the color value of the extracted image, which reflects the degree of image color presentation within the extracted image area MM; MY is the image color value, and MY reflects the overall image background presentation degree within the total image area ZM; L is the image brightness value; The extraction process of the image feature preliminary extraction processing unit is as follows: Through the image acquisition module, key features including image area ratio MB, edge complexity value BF, color contrast value Y and image brightness value L are collected and preliminarily extracted, and after being input into the image feature preliminary extraction processing unit for processing, an image feature value TC for evaluating the significance and quality of the bearing processing area in the image data is output.
4. The industrial intelligent robot according to claim 3, characterized in that: The calculation formula of the fine processing parameter adjustment unit is as follows: ; in: TZ is the adjusted fine parameter value, including cutting speed and feed rate; BJ is the standard processing parameter value; a is a natural value, 1+a -TC*Y It reflects the change in sensitivity of the fine parameter value TZ after adjustment when the image characteristic value TC changes; The adjustment process of the fine processing parameter adjustment unit is as follows: Based on the image feature value TC output by the image feature preliminary extraction processing unit, combined with the standard processing parameter value BJ, and the sensitivity change of the adjusted fine parameter value TZ when the image feature value TC changes, the adjusted fine parameter value TZ adapted to the current image feature is output.
5. The industrial intelligent robot according to claim 4, characterized in that: The calculation formula of the image intelligent adjustment optimization unit is as follows: ; in: MB new is the area ratio of the adjusted image; b is the adjustment factor; BTZ is the target fine parameter value; It is used to feedback and adjust the image area ratio MB; The adjustment and optimization process of the image intelligent adjustment and optimization unit is as follows: Based on the image feature value TC output by the image feature preliminary extraction processing unit and the adjusted fine parameter value TZ output by the fine processing parameter adjustment unit, and combined with the target fine parameter value BTZ, the adjustment factor b is adjusted to output the adjusted image area ratio MB that affects the next intelligent robot processing direction recognition. new .
6. The industrial intelligent robot according to claim 5, characterized in that: Based on the adjusted image area ratio MB output by the image intelligent adjustment optimization unit new The recognition adjustment is as follows: The adjusted image area ratio MB output by the image intelligent adjustment optimization unit at any time new , are used as the input value in the next image feature preliminary extraction processing unit, that is, the image area ratio MB. In other words, the adjusted image area ratio MB outputted this time new , replacing the image area ratio MB input next time, and performing iterative processing, so as to continuously identify the adjusted image area ratio MB, so as to enable the industrial robot to identify and process the bearing processing area direction.
7. An industrial intelligent robot according to claim 6, characterized in that: The adjusted image area ratio MB new The control and response of the adjustment range are as follows: S1. Set MB max is the maximum image area ratio, and sets MB min is the minimum ratio of image area; S2, if the image area maximum ratio MB max Greater than the adjusted image area ratio MB new Greater than the minimum image area ratio MB min , then it is considered that the adjusted image area ratio MB new It is within the safe adjustment range and should be maintained and adjusted continuously. S3, if the adjusted image area ratio MB new Greater than the maximum image area ratio MB max , if the adjusted image area ratio MB new Smaller than the minimum image area ratio MB min Regardless of the above situation, the adjusted image area ratio MB new When in a dangerous adjustment range, the sensor monitors and issues a warning, stops adjustment, and takes action.
8. The industrial intelligent robot according to claim 3, characterized in that: The edge complexity value BF specifically adopts edge detection algorithms, including Canny and SOBEL to identify edges in image data, and detects edges by calculating the gradient strength and direction of pixels in the image data and their neighborhood pixels, extracts features from the detected edges, and the features include the number, length, curvature, and direction of the edges. The features are used to quantify the complexity of the edges, and outputs the edge complexity value BF using algorithms including weighted summation and principal component analysis.
Citation Information
Patent Citations
Image processing method and device, equipment and medium
CN116523932A
Storage safety intelligent monitoring method and system based on image recognition
CN117238100A
Intelligent machining control system for net rack rod piece
CN117961382A
Dam leakage area identification method and device based on unmanned aerial vehicle remote sensing infrared thermal image color distance
CN118115900A
Industrial product appearance detection system and detection method based on AI artificial intelligence visual processing
CN118464906A
