Chip image recognition classification system and method thereof

Abstract

The application discloses a chip image recognition and classification system and a method thereof, and belongs to the technical field of chip recognition. The system comprises a conveying belt, a chip image classification and recognition device, a fixing mechanism and a supporting plate. The supporting plate is arranged on one side of the conveying belt, a first classification box is arranged on the other side of the conveying belt, and a second classification box is arranged at the tail end of the conveying belt. The fixing mechanism is arranged on the supporting plate and used for adjusting the height of a mounting plate relative to the conveying belt. The chip image classification and recognition device is fixed to the side of the mounting plate facing the conveying belt. The chip image classification and recognition device is internally provided with an acquisition device and a control device. The acquisition device is used for acquiring image data of chips on the conveying belt and is connected with an upper computer. A data processing module is arranged in the upper computer. The image data of the chips are preprocessed, features are extracted, and the results after classification and recognition are output to the control device. The control device is used for controlling an electric push rod or the conveying belt, so that the chips are classified and stored.

Description

Technical Field

[0001] This invention belongs to the field of chip recognition technology, specifically relating to a chip image recognition and classification system and method. Background Technology

[0002] Computer vision is a technical field that studies how to enable machines to simulate the human visual system and understand images and videos. In chip image recognition and classification systems, computer vision technology is applied to tasks such as image acquisition, preprocessing, feature extraction, and image classification. While existing technologies can identify chips, they cannot automatically classify them. Therefore, there is an urgent need to provide a system capable of chip image recognition and chip classification based on the recognition results, in order to improve the practicality of the equipment and better meet the needs of real-world applications. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art and to provide a chip image recognition and classification system and method.

[0004] The specific technical solution adopted in this invention is as follows:

[0005] In a first aspect, the present invention provides a chip image recognition and classification system, comprising a conveyor belt, a chip image classification and recognition device, a fixing mechanism, and a support plate. The support plate is disposed on one side of the conveyor belt, and a first classification box is disposed on the other side of the conveyor belt, with the line connecting the support plate and the first classification box perpendicular to the conveying direction of the conveyor belt. A second classification box is disposed at the end of the conveyor belt. A mounting groove for mounting the fixing mechanism is formed at the upper end of the support plate. The fixing mechanism is connected to one end of the mounting plate. The chip image classification and recognition device is fixed to the other end of the mounting plate facing the conveyor belt. By adjusting the vertical position of the mounting plate within the fixing mechanism, the height of the chip image classification and recognition device relative to the conveyor belt is adjusted. An electric push rod is disposed on the support plate, and a push plate for pushing chips on the conveyor belt into the first classification box is fixed to the output end of the electric push rod.

[0006] The chip image classification and recognition equipment includes an acquisition device and a control device. The acquisition device acquires image data of the chips on the conveyor belt and connects to a host computer. The host computer has a data processing module that preprocesses the chip image data, extracts features, and outputs the classification and recognition results to the control device. The control device then controls the electric push rod or the conveyor belt.

[0007] Preferably, the fixing mechanism includes a guide rod and a fixing assembly. The guide rod is vertically fixed in the mounting groove, and one end of the mounting plate is fitted onto the guide rod. The fixing assembly is located on the side wall of the mounting plate and includes a fixing rod, a limiting plate, and an elastic element. One end of the elastic element is fixed to the side wall of the mounting plate, and the other end of the elastic element is fixedly connected to one side of the limiting plate. The fixing rod is fixed to the other side of the limiting plate. Several fixing holes for placing the fixing rod are opened vertically in the mounting groove. By adjusting the position of the fixing rod in the fixing holes, the height of the mounting plate relative to the conveyor belt is changed.

[0008] Preferably, the data acquisition device is a camera or a scanner.

[0009] Preferably, the control device uses a programmable logic controller.

[0010] Preferably, the data processing module uses a residual neural network (ResNet) to take the chip's image data as input and output the chip's classification results.

[0011] Preferably, the elastic element is a spring.

[0012] Secondly, the present invention provides a method for utilizing the chip image recognition and classification system described in the first aspect, as follows:

[0013] S1: Place the chip that needs to be identified and classified on the conveyor belt for transportation. When the chip is transported directly below the chip image classification and recognition device, the control device controls the conveyor belt to stop.

[0014] S2: The acquisition device in the chip image classification and recognition equipment acquires image data from the chip and transmits it to the data processing module in the host computer. Preprocessing is performed using a processing model to improve image clarity, thereby enhancing the accuracy of subsequent image recognition. The preprocessed image data is then used to extract features that aid in recognition and classification. Based on these extracted features, a deep learning algorithm in the classification model classifies the image data, and the classification result is output to the control device.

[0015] S3: If the chip classification result meets the requirements, the control device controls the electric push rod to start, pushing the chip into the first classification box via the push plate. Otherwise, the control device controls the conveyor belt to start, transporting the chip to the second classification box, completing the chip classification and identification.

[0016] Compared with the prior art, the present invention has the following advantages:

[0017] (1) The chip image recognition and classification system provided by this invention has both chip recognition and chip classification functions. Image data of the chip is acquired through an acquisition device and then transmitted to a data processing module in a host computer. Preprocessing is performed using a processing model, followed by the extraction of features that aid in recognition and classification. Based on the extracted features, the image data is classified using a deep learning algorithm in the classification model to obtain the classification result. Finally, the control device transports the chip to different classification bins according to the classification result, thus achieving classified storage of the chip.

[0018] (2) The chip image classification and recognition device provided by the present invention is fixed on a mounting plate. The height of the mounting plate relative to the conveyor belt can be adjusted by a fixing mechanism to accommodate different chip types and sizes. In this way, the acquisition device in the chip image classification and recognition device can be adjusted accordingly to ensure the recognition and classification effect of the system. Attached Figure Description

[0019] Figure 1 Flowchart of the chip image recognition and classification system provided by the present invention;

[0020] Figure 2 A schematic diagram of the chip image recognition and classification system provided in this embodiment;

[0021] Figure 3 A schematic diagram showing the connection between the fixing mechanism and the mounting plate provided in the embodiment;

[0022] Figure 4 for Figure 3 A magnified view of a portion of the fixed component;

[0023] In the diagram: 1. Conveyor belt; 2. Push plate; 3. First sorting box; 4. Second sorting box; 5. Chip image classification and recognition equipment; 6. Mounting plate; 7. Fixing mechanism; 71. Mounting groove; 72. Guide rod; 73. Fixing component; 731. Fixing rod; 732. Limiting plate; 733. Elastic element; 74. Fixing hole; 8. Support plate; 9. Electric push rod. Detailed Implementation

[0024] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below. Technical features in the various embodiments of the present invention can be combined accordingly without mutual conflict.

[0025] In the description of this invention, it should be understood that the terms "first" and "second" are used only for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first" and "second" may explicitly or implicitly include at least one of those features.

[0026] This embodiment, as a preferred embodiment of the present invention, specifically provides a chip image recognition and classification system, including a conveyor belt 1, a chip image classification and recognition device 5, a fixing mechanism 7, and a support plate 8, as detailed below. Figure 2 As shown.

[0027] A support plate 8 is disposed on one side of the conveyor belt 1, and a first sorting box 3 is disposed on the other side of the conveyor belt 1. Both the support plate 8 and the first sorting box 3 are located in the middle of the conveyor belt 1, and the line connecting the support plate 8 and the first sorting box 3 is perpendicular to the conveying direction of the conveyor belt 1. A second sorting box 4 is disposed at the end of the conveyor belt 1.

[0028] The upper end of the support plate 8 has a mounting groove 71 for setting the fixing mechanism 7; the fixing mechanism 7 is connected to one end of the mounting plate 6; the chip image classification and recognition device 5 is fixed on the other end of the mounting plate 6 facing the conveyor belt 1. By adjusting the vertical position of the mounting plate 6 in the fixing mechanism 7, the height of the chip image classification and recognition device 5 relative to the conveyor belt 1 can be adjusted.

[0029] Figure 3 This is a schematic diagram showing the connection between the fixing mechanism and the mounting plate. (See diagram below.) Figure 3 As shown, the fixing mechanism 7 includes a guide rod 72 and a fixing component 73. The fixing component 73 is fixedly connected to the side wall of the mounting plate 6, and includes a fixing rod 731, a limiting plate 732, and an elastic element 733, as detailed below. Figure 4 As shown.

[0030] The guide rod 72 in the fixing mechanism is vertically fixed in the mounting groove 71. A hole is formed at one end of the mounting plate 6, through which the guide rod 72 passes, and the mounting plate 6 is fitted onto the guide rod 72 through the hole. One end of the elastic element 733 is fixed to the side wall of the mounting plate 6, and the other end of the elastic element 733 is fixedly connected to one side of the limiting plate 732. In this embodiment, the elastic element 733 is a spring, but other deformable elastic materials can also be used. A fixing rod 731 is fixed to the other side of the limiting plate 732. Several fixing holes 74 for placing the fixing rod 731 are formed vertically in the mounting groove 71.

[0031] In this embodiment, the fixing rod 731 is a cylinder, and the cross-section of the fixing hole 74 matches that of the fixing rod 731. By pressing one end of the fixing rod 731 to compress the limiting plate 732, the elastic element 733 is compressed, changing the height of the mounting plate 6 relative to the conveyor belt 1. When the set height of the mounting plate 6 is determined, the fixing rod 731 is inserted into the fixing hole 74 at the corresponding height, and the elastic element 733 rebounds, thus limiting and fixing the mounting plate 6. This method can improve the stability of the mounting plate 6 during its up-and-down movement.

[0032] By changing the relative height of the mounting plate 6 to the conveyor belt 1, the height of the chip image classification and recognition device 5 relative to the conveyor belt 1 is changed, enabling the acquisition device to adapt to the image acquisition of different types of chips.

[0033] like Figure 2 As shown, an electric push rod 9 is provided on the support plate 8, and a push plate 2 is fixed to the output end of the electric push rod 9. The electric push rod 9 drives the push rod 2 to move in a direction perpendicular to the movement direction of the chip on the conveyor belt 1. The push plate 2 can push the chip on the conveyor belt 1 into the first sorting box 3.

[0034] The flowchart of the chip image recognition and classification system provided in this embodiment is as follows: Figure 1 As shown, the chip image classification and recognition device 5 has an internal acquisition device and a control device. In this embodiment, the acquisition device is a camera or scanner, used to acquire image data of the chips on the conveyor belt 1 and connect to the host computer.

[0035] The host computer includes a data processing module, which contains a processing model, an extraction model, and a classification model. In this embodiment, the data processing module uses a ResNet residual neural network. The ResNet acquires image data of the chip captured by the acquisition device and preprocesses the image data using the processing model, such as cropping, translating, and noise reduction, thereby improving the clarity of the image data. The extraction model extracts features that are helpful for identification and classification from the preprocessed image data. Based on the extracted features, the chip image is classified using the classification model to obtain the classification result. The classification model uses a deep learning algorithm. The host computer outputs the classification result to the control device, which controls the electric push rod 9 or the conveyor belt 1. In this embodiment, the control device uses a programmable logic controller (PLC). The PLC can receive signals from the host computer, process them through logic operations and control algorithms, and then control the electric push rod 9 or the conveyor belt 1 through the output module.

[0036] This embodiment also provides a method for chip recognition and classification using the chip image recognition and classification system, as follows:

[0037] S1: Place the chip that needs to be identified and classified on the conveyor belt 1 for transportation. When the chip is transported to the chip image classification and recognition device 5, the PLC controls the conveyor belt 1 to stop.

[0038] S2: The acquisition device in the chip image classification and recognition device 5 acquires image data from the chip and transmits the acquired image data to the data processing module in the host computer. The data is preprocessed by the processing model to improve the image clarity, thereby improving the accuracy of subsequent image recognition. The preprocessed image data is then extracted by the extraction model to extract features that are helpful for recognition and classification. Based on the extracted features, the image data is classified by the deep learning algorithm in the classification model, and the classification result is output to the PLC.

[0039] S3: If the chip classification result meets the requirements, the PLC controls the electric push rod 9 to start, and pushes the chip into the first classification box 3 through the push plate 2; if the chip classification result does not meet the requirements, the PLC controls the conveyor belt 1 to start, and transports the chip to the second classification box 4 to complete the chip classification and identification.

[0040] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the invention. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, all technical solutions obtained through equivalent substitution or transformation fall within the protection scope of the present invention.

Claims

1. A chip image recognition and classification system, characterized in that, The system includes a conveyor belt (1), a chip image classification and recognition device (5), a fixing mechanism (7), and a support plate (8). The support plate (8) is located on one side of the conveyor belt (1), and a first classification box (3) is located on the other side of the conveyor belt (1). The line connecting the support plate (8) and the first classification box (3) is perpendicular to the conveying direction of the conveyor belt (1). A second classification box (4) is located at the end of the conveyor belt (1). An installation groove (71) for mounting the fixing mechanism (7) is provided on the upper end of the support plate (8). The fixing mechanism (7) is connected to one end of the mounting plate (6); the other end of the mounting plate (6) facing the conveyor belt (1) is fixed with a chip image classification and recognition device (5). By adjusting the vertical position of the mounting plate (6) in the fixing mechanism (7), the height of the chip image classification and recognition device (5) relative to the conveyor belt (1) can be adjusted; the support plate (8) is provided with an electric push rod (9), and the output end of the electric push rod (9) is fixed with a push plate (2) that pushes the chips on the conveyor belt (1) into the first classification box (3); The chip image classification and recognition device (5) is equipped with an acquisition device and a control device. The acquisition device is used to acquire image data of the chip on the conveyor belt (1) and connect it to the host computer. The host computer is equipped with a data processing module, which outputs the results of preprocessing, extracting features, and classifying the chip image data to the control device. The control device controls the electric push rod (9) or the conveyor belt (1).

2. The chip image recognition and classification system according to claim 1, characterized in that, The fixing mechanism (7) includes a guide rod (72) and a fixing component (73) for adjusting the height of the mounting plate (6) relative to the conveyor belt (1); the guide rod (72) is vertically fixed in the mounting groove (71), and one end of the mounting plate (6) is sleeved on the guide rod (72); the fixing component (73) is disposed on the side wall of the mounting plate (6).

3. The chip image recognition and classification system according to claim 2, characterized in that, The fixing mechanism (7) includes a fixing rod (731), a limiting plate (732), and an elastic element (733); one end of the elastic element (733) is fixed to the side wall of the mounting plate (6), and the other end of the elastic element (733) is fixedly connected to one side of the limiting plate (732); the other side of the limiting plate (732) is fixed with a fixing rod (731); a plurality of fixing holes (74) for placing the fixing rod (731) are opened in the vertical direction on the mounting groove (71), and the height of the mounting plate (6) relative to the conveyor belt (1) is changed by adjusting the position of the fixing rod (731) in the fixing hole (74).

4. The chip image recognition and classification system according to claim 3, characterized in that, The elastic element (733) is a spring.

5. The chip image recognition and classification system according to claim 1, characterized in that, The data acquisition device is a camera or a scanner.

6. The chip image recognition and classification system according to claim 1, characterized in that, The control device employs a programmable logic controller.

7. The chip image recognition and classification system according to claim 1, characterized in that, The data processing module uses a residual neural network (ResNet) to take the chip's image data as input and output the chip's classification results.

8. The chip image recognition and classification system according to claim 1, characterized in that, Both the fixing mechanism (7) and the support plate (8) are made of stainless steel.

9. A method using the chip image recognition and classification system of claim 1, characterized in that, Specifically as follows: S1: The chip that needs to be identified and classified is placed on the conveyor belt (1) for transportation. When the chip is transported to the chip image classification and recognition device (5) directly below, the control device controls the conveyor belt (1) to stop. S2: The acquisition device in the chip image classification and recognition device (5) acquires the image data of the chip and sends the acquired image data to the data processing module in the host computer. The data is preprocessed by the processing model to improve the clarity of the image, thereby improving the accuracy of subsequent image recognition. The preprocessed image data is extracted by the extraction model to extract features that are helpful for recognition and classification. Based on the extracted features, the image data is classified by the deep learning algorithm in the classification model, and the classification result is output to the control device. S3: If the chip classification result meets the requirements, the control device controls the electric push rod (9) to start, and pushes the chip into the first classification box (3) through the push plate (2); otherwise, the control device controls the conveyor belt (1) to start, and transports the chip to the second classification box (4) to complete the chip classification and identification.

Images