Industrial caliper reading recognition method, device and computer readable storage medium

Abstract

An industrial caliper reading recognition method, the method comprises: acquiring an original image of a target device through an image acquisition device; identifying the original image according to a neural network model and a pixel constraint algorithm to obtain a scale image; performing a pixel constraint algorithm on the scale image, analyzing the scale pointed by the caliper arrow, and outputting a reading according to the scale. The application also provides a device and a computer readable storage medium. The application can realize automatic recognition of industrial caliper readings, has strong anti-interference, minimal influence of different illumination angle transformation on the recognition result, high recognition accuracy and speed, can realize full automation without manual operation, automatically stores historical data, and can long-term monitor degree abnormalities.

Description

Technical Field

[0001] The present invention relates to industrial caliper reading recognition technology, and in particular to an industrial caliper reading recognition method, apparatus and computer-readable storage medium. Background Technology

[0002] Large stamping presses and other industrial equipment are typically equipped with various industrial calipers to record the equipment's operating status and material consumption. Caliper readings for these devices must be checked and recorded to monitor whether the readings are within the normal range and condition.

[0003] In existing technologies, since calipers are simple mechanical structures without electronic control systems, most of the inspection work is still done manually by visual inspection, which presents the following problems:

[0004] First, manual visual inspection has a high probability of false detection. After manually recording the inspection form, it needs to be entered into the electronic system again for storage, which is labor-intensive, time-consuming and inefficient.

[0005] Secondly, industrial production scenarios often involve a large number of equipment and production lines, accompanied by unfriendly working environments such as noise and irritating odors. Therefore, there is a possibility that the health of inspection personnel may be affected, which does not meet the needs of green and energy-saving industrial development. For simple mechanical equipment and the reading recognition of old equipment, digital and intelligent transformation is urgently needed to meet the needs of the ever-evolving smart factory scenario. Summary of the Invention

[0006] In view of the above, it is necessary to provide an industrial caliper reading recognition method, a router, and a computer-readable storage medium that can solve many of the defects of the existing technology that uses manual visual inspection.

[0007] This invention provides a method for recognizing industrial caliper readings. The method includes: acquiring an original image of a target device using an image acquisition device; recognizing the original image using a neural network model and a pixel constraint algorithm to obtain a scale image; performing a pixel constraint algorithm on the scale image to parse the scale indicated by the caliper arrow, and outputting a reading based on the scale.

[0008] Optionally, the step of identifying the original image based on the neural network model and pixel constraint algorithm to obtain the scale image includes: achieving initial positioning of the target device based on the neural network model and cropping an effective image; and implementing the pixel constraint algorithm based on the cropped effective image to simulate scale lines and obtain the scale image.

[0009] Optionally, the step of performing a pixel constraint algorithm on the scale image, parsing the scale indicated by the caliper arrow, and outputting a reading based on the scale includes: preprocessing the cropped scale image; calculating the arrow pixel shape; determining whether the arrow pixel shape satisfies feature constraints; when the pixel shape satisfies feature constraints, performing grayscale processing on the scale image, calculating the scale indicated by the arrow, and outputting a reading based on the scale; when the pixel shape does not satisfy feature constraints, recording error information.

[0010] Optionally, the preprocessing of the captured valid image includes: performing noise reduction preprocessing on the captured valid image.

[0011] Optionally, the method further includes saving the original image to a database.

[0012] This invention also provides an apparatus, which includes a memory, a processor, and an industrial caliper reading recognition program stored in the memory and executable on the processor. When executed by the processor, the industrial caliper reading recognition program performs the following steps: acquiring an original image of the target device through an image acquisition device; recognizing the original image according to a neural network model and a pixel constraint algorithm to obtain a scale image; performing a pixel constraint algorithm on the scale image to parse the scale indicated by the caliper arrow, and outputting a reading according to the scale.

[0013] Optionally, the step of identifying the original image based on the neural network model and pixel constraint algorithm to obtain the scale image includes: achieving initial positioning of the target device based on the neural network model and cropping an effective image; and implementing the pixel constraint algorithm based on the cropped effective image to simulate scale lines and obtain the scale image.

[0014] Optionally, the step of performing a pixel constraint algorithm on the scale image, parsing the scale indicated by the caliper arrow, and outputting a reading based on the scale includes: preprocessing the cropped scale image; calculating the arrow pixel shape; determining whether the arrow pixel shape satisfies feature constraints; when the pixel shape satisfies feature constraints, performing grayscale processing on the scale image, calculating the scale indicated by the arrow, and outputting a reading based on the scale; when the pixel shape does not satisfy feature constraints, recording error information.

[0015] Optionally, the preprocessing of the captured valid image includes: performing noise reduction preprocessing on the captured valid image.

[0016] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the industrial caliper reading recognition method described above.

[0017] Compared with existing technologies, the bandwidth dynamic allocation method, router, and computer-readable storage medium described herein can automatically identify industrial caliper readings. It has the advantages of strong anti-interference ability, minimal impact of different lighting angle changes on the identification results, high identification accuracy and speed, full automation without manual operation, automatic storage of historical data, and long-term monitoring of degree anomalies. Attached Figure Description

[0018] Figure 1 This is a diagram illustrating the operating environment of a device according to a preferred embodiment of the present invention.

[0019] Figure 2 This is a program module diagram of a preferred embodiment of the industrial caliper reading recognition system of the present invention.

[0020] Figure 3 This is a flowchart of a preferred embodiment of the industrial caliper reading identification method of the present invention.

[0021] Figure 4 This is a schematic diagram of the original image, effective image, and scale image of a preferred embodiment of the present invention.

[0022] Figure 5 This is a schematic diagram of the arrow pixel distribution pattern of a preferred embodiment of the present invention.

[0023] Figure 6 This is a schematic diagram of a preferred embodiment of the present invention, which obtains scale readings by threading a needle.

[0024] Explanation of main component symbols

[0025] Device 1 Industrial caliper reading recognition system 10 memory 20 processor 30 Data Acquisition Module 101 Recognition Module 102 Output module 103 Save module 104 step S300~S306 Detailed Implementation

[0026] See Figure 1 The diagram shown illustrates the operating environment of a preferred embodiment of the apparatus according to the present invention. The apparatus 1 includes an operating industrial caliper reading recognition system 10. The apparatus 1 also includes a memory 20 and a processor 30, etc.

[0027] The memory 20 includes at least one type of readable storage medium, such as flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. The processor 30 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip.

[0028] See Figure 2 The diagram shown is a program block diagram of a preferred embodiment of the industrial caliper reading recognition system 10 of the present invention.

[0029] The industrial caliper reading recognition system 10 includes a data acquisition module 101, a recognition module 102, an output module 103, and a storage module 104. The modules are configured to be executed by one or more processors (in this embodiment, one processor 30) to complete the invention. A module, as referred to in this invention, is a computer program segment that executes a specific instruction. The memory 20 is used to store the program code and other data of the industrial caliper reading recognition system 10. The processor 30 is used to execute the program code stored in the memory 20.

[0030] The acquisition module 101 is used to acquire raw images of the target device through an image acquisition device.

[0031] In this embodiment, the image acquisition device can be an industrial mobile phone, camera, or webcam. The target device is the device from which readings need to be recorded. The original image, such as... Figure 4 As shown in Figure A.

[0032] The recognition module 102 is used to recognize the original image based on the neural network model and pixel constraint algorithm to obtain the scale image.

[0033] Specifically, the recognition module 102 achieves initial positioning of the target device based on the neural network model and captures valid images; based on the captured valid images, a pixel constraint algorithm is implemented to simulate scale lines and obtain scale images.

[0034] In this embodiment, a valid image refers to an image with scale markings. For example... Figure 4 As shown in B. The scale image, as shown... Figure 4 As shown in C.

[0035] The output module 103 is used to perform a pixel constraint algorithm on the scale image, parse the scale indicated by the caliper arrow, and output a reading based on the scale.

[0036] Specifically, the output module 103 first preprocesses the captured scale image, specifically by performing noise reduction preprocessing on the captured valid image; then, it calculates the arrow pixel shape and determines whether the arrow pixel shape satisfies the feature constraints; finally, when the pixel shape satisfies the feature constraints, it performs grayscale processing on the scale image, calculates the scale indicated by the arrow, and outputs a reading based on the scale; when the pixel shape does not satisfy the feature constraints, it records an error message.

[0037] In this embodiment, when the image of the caliper arrow is read, the numerical matrix of the scale image is obtained through noise reduction and grayscale conversion. Then, different pixel distribution characteristics are obtained by calculating the pixel values. For example... Figure 5 As shown, taking the features of an arrow image as an example, it should be a convex peak shape with a high center and low ends. Based on this, the location is determined to be the arrow position, and this feature is regarded as the pixel constraint for arrow positioning.

[0038] In this embodiment, the pixel sequence of the caliper scale readings exhibits a wavy pattern. By combining this with the threading method to determine the number of intersections, the caliper reading can be calculated. Therefore, as... Figure 6 As shown, once the position of the arrow is confirmed, the caliper reading can be calculated using the needle-threading method.

[0039] The storage module 104 is used to save caliper readings to the database.

[0040] In this embodiment, the storage module 104 saves the caliper readings to a designated database or electronic archive. This enables automatic saving of historical data and long-term monitoring of reading anomalies.

[0041] In this embodiment, automatic recognition of industrial caliper readings can be achieved. It has the advantages of strong anti-interference ability, minimal impact of different lighting angles on the recognition results, high recognition accuracy and speed, full automation without manual operation, automatic storage of historical data, and long-term monitoring of abnormal readings.

[0042] See Figure 3 The diagram shown is a flowchart of a preferred embodiment of the industrial caliper reading recognition method of the present invention. The industrial caliper reading recognition method is applied to device 1 and can be executed by the processor 30. Figure 2 It is implemented by modules 101 to 104 shown.

[0043] Step S300: Acquire the original image of the target device using the image acquisition device.

[0044] In this embodiment, the image acquisition device can be an industrial mobile phone, camera, or webcam. The target device is the device from which readings need to be recorded. The original image, such as... Figure 4 As shown in Figure A.

[0045] Step S302: The original image is identified according to the neural network model and pixel constraint algorithm to obtain the scale image.

[0046] Specifically, the initial positioning of the target device is achieved based on the neural network model, and a valid image is captured; based on the captured valid image, a pixel constraint algorithm is implemented to simulate the scale lines and obtain the scale image.

[0047] In this embodiment, a valid image refers to an image with scale markings. For example... Figure 4 As shown in B. The scale image, as shown... Figure 4 As shown in C.

[0048] Step S304: Perform a pixel constraint algorithm on the scale image, parse the scale indicated by the caliper arrow, and output the reading based on the scale.

[0049] Specifically, firstly, the captured scale image is preprocessed, specifically, the captured valid image is subjected to noise reduction preprocessing; then, the arrow pixel shape is calculated, and it is determined whether the arrow pixel shape satisfies the feature constraints; finally, when the pixel shape satisfies the feature constraints, the scale image is grayscaled, the scale indicated by the arrow is calculated, and the reading is output according to the scale; when the pixel shape does not satisfy the feature constraints, error information is recorded.

[0050] In this embodiment, when the image of the caliper arrow is read, the numerical matrix of the scale image is obtained through noise reduction and grayscale conversion. Then, different pixel distribution characteristics are obtained by calculating the pixel values. For example... Figure 5 As shown, taking the features of an arrow image as an example, it should be a convex peak shape with a high center and low ends. Based on this, the location is determined to be the arrow position, and this feature is regarded as the pixel constraint for arrow positioning.

[0051] In this embodiment, the pixel sequence of the caliper scale readings exhibits a wavy pattern. By combining this with the threading method to determine the number of intersections, the caliper reading can be calculated. Therefore, as... Figure 6 As shown, once the position of the arrow is confirmed, the caliper reading can be calculated using the needle-threading method.

[0052] Step S306: Save the caliper readings to the database.

[0053] In this embodiment, caliper readings are saved to a designated database or electronic archive. This allows for automatic saving of historical data and long-term monitoring of reading anomalies.

[0054] By applying the above method to the above device, automatic recognition of industrial caliper readings can be achieved. It has the advantages of strong anti-interference, minimal impact of different lighting angles on the recognition results, high recognition accuracy and speed, full automation without manual operation, automatic storage of historical data, and long-term monitoring of abnormal readings.

[0055] It is worth noting that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for identifying readings of industrial calipers, characterized in that, The method includes: The original image of the target device is acquired using an image acquisition device; The original image is identified using a neural network model and a pixel constraint algorithm to obtain a scale image; A pixel constraint algorithm is applied to the scale image to parse the scale indicated by the caliper arrow, and a reading is output based on the scale. The step of identifying the original image based on a neural network model and a pixel constraint algorithm to obtain a scale image includes: The target device is initially located based on a neural network model, and valid images are captured. Based on the captured valid image, a pixel constraint algorithm is implemented to simulate the scale lines and obtain the scale image. The step of performing a pixel constraint algorithm on the scale image, parsing the scale indicated by the caliper arrow, and outputting a reading based on the scale includes: Preprocess the extracted scale image; Calculate the arrow pixel shape; Determine whether the shape of the arrow pixels satisfies the feature constraints; When the pixel shape satisfies the feature constraints, the scale image is converted to grayscale, the scale indicated by the arrow is calculated, and the reading is output according to the scale. When the pixel shape does not meet the feature constraints, an error message is recorded; The numerical matrix of the scale image is obtained through noise reduction and grayscale conversion. Then, multiple pixel distribution features are obtained by calculating the pixel values. Among the multiple pixel distribution features, the pixel distribution feature with a convex peak shape that is high in the middle and low at both ends is the pixel constraint for arrow positioning. The location region of the pixel distribution feature with a convex peak shape that is high in the middle and low at both ends is the arrow position.

2. The industrial caliper reading identification method as described in claim 1, characterized in that, The preprocessing of the captured valid image includes: The captured valid image is subjected to noise reduction preprocessing.

3. The industrial caliper reading identification method as described in claim 1, characterized in that, The method further includes: Save caliper readings to the database.

4. An apparatus, characterized in that, The device includes a memory, a processor, and an industrial caliper reading recognition program stored in the memory and executable on the processor. When the industrial caliper reading recognition program is executed by the processor, it performs the following steps: The original image of the target device is acquired using an image acquisition device; The original image is identified using a neural network model and a pixel constraint algorithm to obtain a scale image; A pixel constraint algorithm is applied to the scale image to parse the scale indicated by the caliper arrow, and a reading is output based on the scale. The step of identifying the original image based on a neural network model and a pixel constraint algorithm to obtain a scale image includes: The target device is initially located based on a neural network model, and valid images are captured. Based on the captured valid image, a pixel constraint algorithm is implemented to simulate the scale lines and obtain the scale image. The step of performing a pixel constraint algorithm on the scale image, parsing the scale indicated by the caliper arrow, and outputting a reading based on the scale includes: Preprocess the extracted scale image; Calculate the arrow pixel shape; Determine whether the shape of the arrow pixels satisfies the feature constraints; When the pixel shape satisfies the feature constraints, the scale image is converted to grayscale, the scale indicated by the arrow is calculated, and the reading is output according to the scale. When the pixel shape does not meet the feature constraints, an error message is recorded; The numerical matrix of the scale image is obtained through noise reduction and grayscale conversion. Then, multiple pixel distribution features are obtained by calculating the pixel values. Among the multiple pixel distribution features, the pixel distribution feature with a convex peak shape that is high in the middle and low at both ends is the pixel constraint for arrow positioning. The location region of the pixel distribution feature with a convex peak shape that is high in the middle and low at both ends is the arrow position.

5. The apparatus as described in claim 4, characterized in that, The preprocessing of the captured valid image includes: The captured valid image is subjected to noise reduction preprocessing.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the industrial caliper reading identification method as described in any one of claims 1 to 3.

Images