Automatic detection and evaluation method and system for waste recycled steel

Abstract

The invention relates to the technical field of automatic detection, in particular to an automatic detection and evaluation method for waste recycled steel, which comprises the following steps: executing automatic image acquisition on the waste recycled steel to obtain a recycled steel acquisition image, storing the recycled steel acquisition image into an image storage unit and executing image preprocessing on the recycled steel acquisition image to obtain a preprocessed image; performing fuzzy marking on the preprocessed image to obtain a fuzzy boundary set, sequentially extracting fuzzy boundaries, judging whether the fuzzy boundaries are defect boundaries or not, judging defect types of the defect boundaries, calculating defect scores of the defect boundaries based on the defect types, and if the fuzzy boundaries are extracted, summarizing the defect scores to obtain a defect score set; and calculating a comprehensive defect score based on the defect score set, and evaluating and grading the waste recycled steel based on the score grade and the comprehensive defect score to obtain graded recycled steel. According to the method, the detection and evaluation efficiency of the waste recycled steel can be improved.

Description

Technical Field

[0001] The present invention relates to the field of automated detection technology, and in particular to an automated detection and evaluation method, system, electronic equipment and computer-readable storage medium for scrap recycled steel. Background Art

[0002] With the continuous development of scrap metal recycling, the detection and evaluation of scrap recycled steel has become an important part of improving resource utilization efficiency.

[0003] Existing methods for detecting scrap recycled steel rely on manual inspection, where inspectors visually check the appearance of the recycled steel for defects such as cracks, holes, and rust.

[0004] Although the above methods can detect scrap recycled steel, manual detection is inefficient and easily affected by subjective factors. Secondly, when faced with complex backgrounds and changing lighting conditions, inspectors are prone to misjudgment, affecting the reliability of the evaluation results. In addition, manual detection lacks systematicity in defect classification and evaluation, and cannot effectively quantify different types of defects. Therefore, the current detection and evaluation methods for scrap recycled steel have the problem of low detection and evaluation efficiency. Summary of the invention

[0005] The present invention provides an automated detection and evaluation method for scrap recycled steel and a computer-readable storage medium, the main purpose of which is to improve the detection and evaluation efficiency of scrap recycled steel.

[0006] To achieve the above-mentioned purpose, the present invention provides an automated detection and evaluation method for scrap recycled steel, comprising:

[0007] Receiving a detection instruction, and starting a pre-built image detection unit based on the detection instruction, wherein the image detection unit includes: an image acquisition unit, an image storage unit and an image processing unit;

[0008] Utilizing the image acquisition unit to perform automated image acquisition on the preset scrap recycled steel, obtaining the recycled steel acquisition image and generating an image storage instruction;

[0009] Using the image storage unit to receive the image storage instruction, storing the recycled steel collected image into the image storage unit and generating an image processing instruction;

[0010] Utilizing the image processing unit to receive the image processing instruction and perform processing instruction parsing to obtain an instruction parsing result;

[0011] Based on the instruction analysis result and the image processing unit, image preprocessing is performed on the collected image of the recycled steel to obtain a preprocessed image, wherein the image preprocessing includes: median filtering and binarization;

[0012] Performing fuzzy labeling on the preprocessed image using a preset fuzzy labeling method to obtain a fuzzy boundary set;

[0013] Extracting fuzzy boundaries in sequence from the fuzzy boundary set;

[0014] Using a preset defect threshold to determine whether the fuzzy boundary is a preset defect boundary;

[0015] If the fuzzy boundary is not a defect boundary, returning to the above step of sequentially extracting fuzzy boundaries from the fuzzy boundary set;

[0016] If the fuzzy boundary is a defect boundary, the defect type of the defect boundary is determined based on a preset defect standard, wherein the defect type includes: crack defect, hole defect and corrosion defect;

[0017] Calculating a defect score for the defect boundary based on the defect type;

[0018] Determining whether the fuzzy boundary is extracted;

[0019] If the fuzzy boundary extraction is not completed, returning to the above step of sequentially extracting fuzzy boundaries in the fuzzy boundary set;

[0020] If the fuzzy boundary extraction is completed, the defect scores are summarized to obtain a defect score set, and a comprehensive defect score is calculated based on the defect score set;

[0021] The scrap recycled steel is evaluated and graded based on a preset scoring level and the comprehensive defect score to obtain graded recycled steel.

[0022] Optionally, the using the image acquisition unit to perform automatic image acquisition on the preset scrap recycled steel to obtain the recycled steel acquisition image and generate an image storage instruction includes:

[0023] Utilizing the image acquisition unit to continuously photograph the scrap recycled steel for 3 seconds at a shooting frequency of 5 images / s to obtain a recycled steel image set;

[0024] Based on the recycled steel image set, two images with the highest definition are selected to obtain a first representative image and a second representative image;

[0025] Acquire a first pixel value set of the first representative image and a second pixel value set of the second representative image;

[0026] Expanding the first pixel value set and the second pixel value set into a one-dimensional array to obtain a first pixel array and a second pixel array;

[0027] Calculating the image similarity between the first representative image and the second representative image based on the first pixel array and the second pixel array;

[0028] Determining whether the image similarity is greater than a preset similarity threshold;

[0029] If the image similarity is greater than a similarity threshold, randomly selecting one of the first representative image and the second representative image as a recycled steel collection image, and generating an image storage instruction based on the recycled steel collection image;

[0030] If the image similarity is not greater than the similarity threshold, the first representative image and the second representative image are used as recycled steel collection images, and an image storage instruction is generated based on the recycled steel collection images.

[0031] Optionally, the calculating the image similarity between the first representative image and the second representative image based on the first pixel array and the second pixel array includes:

[0032] Acquire a total number of pixels based on the first pixel array and the second pixel array, wherein the total number of pixels is the total number of pixels of the first pixel array or the total number of pixels of the second pixel array, and the total number of pixels of the first pixel array is equal to the total number of pixels of the second pixel array;

[0033] The image similarity between the first representative image and the second representative image is calculated using the total number of pixels, the first pixel array, and the second pixel array:

[0034]

[0035] Among them, T refers to the image similarity, n refers to the total number of pixels, and Q i Refers to the i-th pixel value in the first pixel array, W i Refers to the i-th pixel value in the second pixel array.

[0036] Optionally, the using the image processing unit to receive the image processing instruction and perform processing instruction parsing to obtain the instruction parsing result includes:

[0037] The image processing unit includes: a first image processing unit and a second image processing unit;

[0038] Processing instruction parsing is as follows:

[0039] Calculating the operating load of the first image processing unit, and determining whether the operating load is less than a preset operating load threshold;

[0040] If the operating load is less than the operating load threshold, determining the instruction analysis result as inputting the recycled steel acquisition image into the first image processing unit;

[0041] If the operating load is equal to or greater than the operating load threshold, the instruction analysis result is determined to input the recycled steel acquisition image into the second image processing unit, and after the operating load is less than the operating load threshold, the recycled steel acquisition image is input into the first image acquisition unit.

[0042] Optionally, the calculating the operating load of the first image processing unit includes:

[0043] Obtaining the CPU usage, memory usage, bandwidth usage and IO rate of the first image processing unit;

[0044] Obtaining a CPU load weight, a memory load weight, a bandwidth load weight, and an IO load weight of the first image processing unit;

[0045] The operation load of the first image processing unit is calculated based on the CPU usage rate, memory usage, bandwidth usage rate, IO rate, CPU load weight, memory load weight, bandwidth load weight and IO load weight:

[0046] H=ρ1×R1+ρ2×R2+ρ3×R3+ρ4×R4

[0047] Among them, H refers to the operating load of the first image processing unit, ρ1 refers to the CPU load weight, R1 refers to the CPU usage rate, ρ2 refers to the memory load weight, R2 refers to the memory usage, ρ3 refers to the bandwidth load weight, R3 refers to the bandwidth usage rate, ρ4 refers to the IO load weight, and R4 refers to the IO rate.

[0048] Optionally, the using a preset defect threshold to determine whether the fuzzy boundary is a defect boundary includes:

[0049] Acquire the total number of fuzzy boundary pixels of the fuzzy boundary and the pixel equivalent of the fuzzy boundary;

[0050] The fuzzy boundary area of ​​the fuzzy boundary is calculated based on the total number of fuzzy boundary pixels of the fuzzy boundary and the pixel equivalent of the fuzzy boundary:

[0051] M=S×Z

[0052] Among them, M refers to the fuzzy boundary area, S refers to the total number of fuzzy boundary pixels, and Z refers to the pixel equivalent;

[0053] Determining whether the fuzzy boundary area is smaller than the defect threshold;

[0054] If the fuzzy boundary area is smaller than the defect threshold, confirming that the fuzzy boundary is not a defect boundary;

[0055] If the fuzzy boundary area is not less than the defect threshold, the fuzzy boundary is confirmed to be a defect boundary.

[0056] Optionally, judging the defect type of the defect boundary based on a preset defect standard includes:

[0057] Obtaining a defect boundary area of ​​the defect boundary, and determining whether the defect boundary area is smaller than a preset crack area threshold;

[0058] If the defect boundary area is smaller than the crack area threshold, confirming that the defect type of the defect boundary is a crack defect;

[0059] If the defect boundary area is not less than the crack area threshold, then calculating the defect boundary roundness of the defect boundary;

[0060] The calculation method is as follows:

[0061] Obtain defective pixels, the total number of defective pixels, and the horizontal and vertical coordinates of the defective pixels at the defect boundary;

[0062] The perimeter of the defect boundary is calculated using the preset dist function, the defect pixels of the defect boundary, the total number of defect pixels, the abscissa and ordinate of the defect pixels:

[0063]

[0064] Among them, X refers to the perimeter of the defect boundary, E refers to the total number of defect pixels, dist refers to the dist function, and x τ Refers to the horizontal coordinate of the τth defective pixel, y τ Refers to the ordinate of the τth defective pixel, x τ+1 Refers to the horizontal coordinate of the τ+1th defective pixel, y τ+1 Refers to the ordinate of the τ+1th defective pixel;

[0065] Calculate the roundness of the defect boundary:

[0066]

[0067] Among them, Y refers to the roundness of the defect boundary, A refers to the area of ​​the defect boundary, X refers to the circumference of the defect boundary, and π refers to pi;

[0068] Determining whether the defect boundary roundness of the defect boundary is less than a preset hole roundness threshold;

[0069] If the defect boundary roundness of the defect boundary is not less than the hole roundness threshold, confirming that the defect type of the defect boundary is a hole defect;

[0070] If the defect boundary roundness of the defect boundary is less than the hole roundness threshold, it is confirmed that the defect type of the defect boundary is a corrosion defect.

[0071] Optionally, the calculating the defect score of the defect boundary based on the defect type includes:

[0072] If the defect type is a crack defect, then obtaining the crack area, crack depth and crack length of the crack defect;

[0073] Substitute the crack area, crack depth and crack length into the preset crack area standard, crack depth standard and crack length standard respectively to obtain the crack area score, crack depth score and crack length score, wherein the crack area standard is: (0,5] cm 2 : 2 points, (5, 10] cm 2 :6 points, (10, +∞)cm 2 : 10 points, the crack depth standard is: (0, 0.5] mm: 2 points, (0.5, 1.5] mm: 6 points, (1.5, +∞) mm: 10 points, the crack length standard is: (0, 2] cm: 2 points, (2, 6] cm: 6 points, (6, +∞) cm: 10 points;

[0074] The defect score of the crack defect is calculated using the crack area score, crack depth score and crack length score:

[0075] U1=b1+b2+b3

[0076] Among them, U1 refers to the defect score of crack defect, b1 refers to the crack area score, b2 refers to the crack depth score, and b3 refers to the crack length score;

[0077] If the defect type is a hole defect, then obtaining the hole diameter, hole depth and hole roundness of the hole defect;

[0078] Substitute the hole diameter, hole depth and hole roundness into the preset hole diameter standard, hole depth standard and hole roundness standard to obtain the hole diameter score, hole depth score and hole roundness score, wherein the hole diameter standard is: (0, 0.5] cm: 2 points, (0.5, 4] cm: 6 points, (4, +∞) cm: 10 points, the hole depth standard is: (0, 0.5] mm: 2 points, (0.5, 3] mm: 6 points, (3, +∞) mm: 10 points, the hole roundness standard is: (0, 0.5]: 10 points, (0.5, 0.8], 6 points, (0.8, 1.0): 2 points;

[0079] The defect score of the hole defect type is calculated using the hole diameter score, hole depth score and hole roundness score:

[0080] U2=g1+g2+g3

[0081] Among them, U2 refers to the defect score of hole defect, g1 refers to the hole diameter score, g2 refers to the hole depth score, and g3 refers to the hole roundness score;

[0082] If the defect type is a rust defect, the rust area, rust depth and texture complexity of the rust defect are obtained, wherein the texture complexity includes: smooth, rough and extremely rough;

[0083] Substitute the rust area, rust depth, and texture complexity into the preset rust area standard, rust depth standard, and texture complexity standard to obtain the rust area score, rust depth score, and texture complexity score, wherein the rust area standard is: (0, 5] cm 2 : 2 points, (5, 10] cm 2 :6 points, (10, +∞) cm 2 : 10 points, the rust depth standard is: (0, 1] mm: 2 points, (1, 4] mm: 6 points, (4, +∞) mm: 10 points, the texture complexity standard is: smooth: 2 points, rough: 6 points, extremely rough: 10 points;

[0084] The defect score of the rust defect type is calculated using the rust area score, rust depth score and texture complexity score:

[0085] U3=k1+k2+k3

[0086] Among them, U3 refers to the defect score of rust defect, k1 refers to the rust area score, k2 refers to the rust depth score, and k3 refers to the texture complexity score.

[0087] Optionally, the calculating a comprehensive defect score based on the defect score set includes:

[0088] Dividing the defect score set into a crack score set, a hole score set and a corrosion score set based on the defect type, and obtaining the number of crack scores, the number of hole scores, the number of corrosion scores, the crack scores, the hole scores and the corrosion scores by using the crack score set, the hole score set and the corrosion score set;

[0089] Obtain crack score weights, hole score weights, and corrosion score weights, and calculate a comprehensive defect score based on the number of crack scores, the number of hole scores, the number of corrosion scores, the crack score, the hole score, the corrosion score, the crack score weights, the hole score weights, and the corrosion score weights:

[0090]

[0091] Among them, U4 refers to the comprehensive defect score, l1 refers to the crack score weight, f refers to the number of crack scores, and h α refers to the αth crack score, l2 refers to the hole score weight, t refers to the number of hole scores, c β refers to the βth hole score, l3 refers to the rust score weight, o refers to the number of rust scores, j γ Refers to the γth rust score.

[0092] To achieve the above object, the present invention also provides an automated detection and evaluation system for recycled scrap steel, comprising:

[0093] An image acquisition module is used to receive a detection instruction and start a pre-built image detection unit based on the detection instruction, wherein the image detection unit includes: an image acquisition unit, an image storage unit and an image processing unit, and the image acquisition unit is used to perform automatic image acquisition on the preset scrap recycled steel to obtain a recycled steel acquisition image and generate an image storage instruction;

[0094] An image preprocessing module is used to receive the image storage instruction by using the image storage unit, store the recycled steel collected image in the image storage unit and generate an image processing instruction, receive the image processing instruction by using the image processing unit and perform processing instruction analysis to obtain an instruction analysis result, and perform image preprocessing on the recycled steel collected image based on the instruction analysis result and the image processing unit to obtain a preprocessed image, wherein the image preprocessing includes: median filtering and binarization;

[0095] a defect boundary judgment module, configured to perform fuzzy marking on the preprocessed image using a preset fuzzy marking method to obtain a fuzzy boundary set, extract fuzzy boundaries in the fuzzy boundary set in sequence, and determine whether the fuzzy boundary is a preset defect boundary using a preset defect threshold; if the fuzzy boundary is not a defect boundary, return to the above step of sequentially extracting fuzzy boundaries in the fuzzy boundary set; if the fuzzy boundary is a defect boundary, determine the defect type of the defect boundary based on a preset defect standard, wherein the defect types include: crack defects, hole defects, and corrosion defects;

[0096] The recycled steel grading module is used to calculate the defect score of the defect boundary based on the defect type, determine whether the fuzzy boundary extraction has been completed, and if the fuzzy boundary extraction has not been completed, return to the above step of extracting fuzzy boundaries in the fuzzy boundary set in sequence; if the fuzzy boundary extraction is completed, summarize the defect scores to obtain a defect score set, and calculate a comprehensive defect score based on the defect score set, and evaluate and grade the scrap recycled steel based on a preset scoring level and the comprehensive defect score.

[0097] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising:

[0098] A memory storing at least one instruction;

[0099] The processor executes the instructions stored in the memory to implement the above-mentioned automatic detection and evaluation method of scrap recycled steel.

[0100] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned automated detection and evaluation method for recycled scrap steel.

[0101] In order to solve the problems described in the background technology, the present invention realizes the automatic detection and evaluation of the scrap recycled steel by combining the scoring level and the comprehensive defect score to perform multi-step processing on the image of the scrap recycled steel; firstly, the image acquisition unit is used to perform automatic image acquisition on the scrap recycled steel. The automatic image acquisition can quickly and continuously obtain the image of the scrap recycled steel, which reduces the waste of manpower and ensures that the real state of the surface of the scrap recycled steel can be captured in time, which provides accurate image information for subsequent processing and improves the efficiency of overall detection and evaluation; secondly, the collected image of the recycled steel is preprocessed by median filtering and binarization. The median filtering and binarization processing can effectively remove background noise and highlight important features, making defects easier to observe in the image, and making fuzzy marking and defect recognition more efficient, thereby improving the accuracy and efficiency of detection; thereafter, the preprocessed image is fuzzy marked by using the fuzzy marking method to obtain a fuzzy boundary set. The fuzzy boundary set can help to more accurately define the defect area, thereby improving the efficiency and accuracy of defect recognition; further, the fuzzy boundary sets are extracted in sequence in the fuzzy boundary set. Boundary, and use the defect threshold to determine whether the extracted fuzzy boundary is a defect boundary. When it is a defect boundary, the defect type of the defect boundary is determined based on the defect standard. Determining the defect type of the defect boundary can reduce subjective judgment and improve the reliability of subsequent defect score calculation; then, the defect score of the defect boundary is calculated based on the defect type, and each defect is scored to quantify the severity of the defect, so that the evaluation result is more intuitive, which not only provides a data basis for evaluation and grading, but also reduces the error of human judgment; then, when the fuzzy boundary is extracted, the defect score is summarized to obtain a defect score set, and a comprehensive defect score is calculated based on the defect score set. The comprehensive defect score is calculated based on the defect score, which can make the quality status of the scrap recycled steel clearer and improve the evaluation efficiency; finally, the scrap recycled steel is evaluated and graded based on the scoring level and the comprehensive defect score, and effective grading is performed based on the comprehensive defect score and the scoring level, which can quickly evaluate the quality of the scrap recycled steel. This standardized evaluation process simplifies the operation process and makes the evaluation of the scrap recycled steel more efficient. Therefore, the present invention can improve the detection and evaluation efficiency of scrap recycled steel. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] Figure 1 A schematic flow chart of an automated detection and evaluation method for scrap recycled steel provided in one embodiment of the present invention;

[0103] Figure 2 A functional module diagram of an automated detection and evaluation system for scrap recycled steel provided by an embodiment of the present invention;

[0104] Figure 3 A schematic diagram of the structure of an electronic device for implementing the automated detection and evaluation method for recycled scrap steel provided in one embodiment of the present invention.

[0105] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0106] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0107] The embodiment of the present application provides an automated detection and evaluation method for scrap recycled steel. The execution subject of the automated detection and evaluation method for scrap recycled steel includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided in the embodiment of the present application. In other words, the automated detection and evaluation method for scrap recycled steel can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0108] Reference Figure 1 FIG. 1 is a flow chart of an automated detection and evaluation method for scrap recycled steel provided by an embodiment of the present invention. In this embodiment, the automated detection and evaluation method for scrap recycled steel includes:

[0109] S1. Receive a detection instruction, and start a pre-built image detection unit based on the detection instruction, wherein the image detection unit includes: an image acquisition unit, an image storage unit and an image processing unit.

[0110] It can be understood that the detection instruction refers to the instruction initiated by the detection personnel for starting the image detection unit. The image detection unit includes an image acquisition unit, an image storage unit and an image processing unit. The image acquisition unit is used to acquire images of scrap recycled steel and generate image storage instructions. The image storage instruction refers to the instruction for storing images, which is sent to the image storage unit through the image acquisition unit. The image storage unit is used to receive the image storage instruction, store the acquired image and generate the image processing instruction. The image processing instruction refers to the instruction for processing the image, which is sent to the image processing unit through the image storage unit. The image processing unit is used to receive the image processing instruction and process the acquired image.

[0111] For example, a scrap steel recycling factory wants to inspect and evaluate a batch of scrap recycled steel. The inspector initiates a detection command and starts the image detection unit. The image acquisition unit begins to automatically capture images of the scrap recycled steel and stores the captured images in the image storage unit. When the image acquisition is completed, the image storage unit sends the captured images to the image processing unit, and the image processing unit processes the images.

[0112] S2. Utilize the image acquisition unit to perform automatic image acquisition on the preset scrap recycled steel, obtain the recycled steel acquisition image and generate an image storage instruction.

[0113] It can be explained that scrap recycled steel refers to steel materials that have been used and discarded in industries such as industry and construction. These steel materials can be reused after recycling. Automated image acquisition refers to the image acquisition unit automatically acquiring images of scrap recycled steel. Recycled steel acquisition images refer to images obtained by the image acquisition unit acquiring images of scrap recycled steel.

[0114] In detail, the method of using the image acquisition unit to perform automatic image acquisition on the preset scrap recycled steel, obtaining the recycled steel acquisition image and generating an image storage instruction includes:

[0115] Utilizing the image acquisition unit to continuously photograph the scrap recycled steel for 3 seconds at a shooting frequency of 5 images / s to obtain a recycled steel image set;

[0116] Based on the recycled steel image set, two images with the highest definition are selected to obtain a first representative image and a second representative image;

[0117] Acquire a first pixel value set of the first representative image and a second pixel value set of the second representative image;

[0118] Expanding the first pixel value set and the second pixel value set into a one-dimensional array to obtain a first pixel array and a second pixel array;

[0119] Calculating image similarity between the first pixel array and the second pixel array;

[0120] Determining whether the image similarity is greater than a preset similarity threshold;

[0121] If the image similarity is greater than a similarity threshold, randomly selecting one of the first representative image and the second representative image as a recycled steel collection image, and generating an image storage instruction based on the recycled steel collection image;

[0122] If the image similarity is not greater than the similarity threshold, the first representative image and the second representative image are used as recycled steel collection images, and an image storage instruction is generated based on the recycled steel collection images.

[0123] It can be explained that the recycled steel image set refers to the set of images formed by shooting the waste recycled steel for 3 seconds at a shooting frequency of 5 images / s, the first representative image and the second representative image refer to the two images with the highest clarity in the recycled steel image set, the first pixel value set refers to the set formed by the pixel values ​​of the first representative image, the second pixel value set refers to the set formed by the pixel values ​​of the second representative image, the first pixel array refers to the array obtained by expanding the first pixel value set into a one-dimensional array, and the second pixel array refers to the array obtained by expanding the second pixel value set into a one-dimensional array. For example, assuming that the first pixel value set is in the form of a 3×3 matrix, Expand the first pixel value set into a one-dimensional array to obtain [1,3,4,2,3,5,7,8,6], where [1,3,4,2,3,5,7,8,6] is the first pixel array. Image similarity refers to the degree of similarity between two images. The similarity threshold refers to a pre-set value used to determine whether two images are similar. When the image similarity is greater than the similarity threshold, the two images are considered to be the same.

[0124] In detail, the calculating the image similarity between the first representative image and the second representative image based on the first pixel array and the second pixel array includes:

[0125] Acquire a total number of pixels based on the first pixel array and the second pixel array, wherein the total number of pixels is the total number of pixels of the first pixel array or the total number of pixels of the second pixel array, and the total number of pixels of the first pixel array is equal to the total number of pixels of the second pixel array;

[0126] The image similarity between the first representative image and the second representative image is calculated using the total number of pixels, the first pixel array, and the second pixel array:

[0127]

[0128] Among them, T refers to the image similarity, n refers to the total number of pixels, and Q i Refers to the i-th pixel value in the first pixel array, W i Refers to the i-th pixel value in the second pixel array.

[0129] It can be understood that the total number of pixels refers to the total number of pixel values ​​in the first pixel array or the second pixel array, and the total number of pixels in the first pixel array is equal to the total number of pixels in the second pixel array.

[0130] S3. Utilize the image storage unit to receive the image storage instruction, store the recycled steel collected image into the image storage unit and generate an image processing instruction.

[0131] S4. Utilize the image processing unit to receive the image processing instruction and perform processing instruction analysis to obtain an instruction analysis result.

[0132] It can be explained that the processing instruction analysis refers to the action of analyzing the image processing instruction, and the instruction analysis result refers to the conclusion obtained after the processing instruction analysis. Through the instruction analysis result, it can be known whether to input the recycled steel collected image directly into the first image processing unit or to input the recycled steel collected image into the second image processing unit first, and then input the recycled steel collected image into the first image processing unit after the operating load of the first image processing unit is less than the operating load threshold. The operating load refers to the load of the first image processing unit when it is working, and the operating load threshold refers to the maximum load that the first image processing unit can withstand when it is working.

[0133] In detail, the using the image processing unit to receive the image processing instruction and perform processing instruction parsing to obtain the instruction parsing result includes:

[0134] The image processing unit includes: a first image processing unit and a second image processing unit;

[0135] Processing instruction parsing is as follows:

[0136] Calculating the operating load of the first image processing unit, and determining whether the operating load is less than a preset operating load threshold;

[0137] If the operating load is less than the operating load threshold, determining the instruction analysis result as inputting the recycled steel acquisition image into the first image processing unit;

[0138] If the operating load is equal to or greater than the operating load threshold, the instruction analysis result is determined to input the recycled steel acquisition image into the second image processing unit, and after the operating load is less than the operating load threshold, the recycled steel acquisition image is input into the first image acquisition unit.

[0139] In detail, the calculating the operation load of the first image processing unit includes:

[0140] Obtaining the CPU usage, memory usage, bandwidth usage and IO rate of the first image processing unit;

[0141] Obtaining a CPU load weight, a memory load weight, a bandwidth load weight, and an IO load weight of the first image processing unit;

[0142] The operation load of the first image processing unit is calculated based on the CPU usage rate, memory usage, bandwidth usage rate, IO rate, CPU load weight, memory load weight, bandwidth load weight and IO load weight:

[0143] H=ρ1×R1+ρ2×R2+ρ3×R3+ρ4×R4

[0144] Among them, H refers to the operating load of the first image processing unit, ρ1 refers to the CPU load weight, R1 refers to the CPU usage rate, ρ2 refers to the memory load weight, R2 refers to the memory usage, ρ3 refers to the bandwidth load weight, R3 refers to the bandwidth usage rate, ρ4 refers to the IO load weight, and R4 refers to the IO rate.

[0145] It can be explained that the CPU utilization rate refers to the degree of CPU utilization when the first image processing unit is working, the memory utilization rate refers to the degree of memory utilization when the first image processing unit is working, the bandwidth utilization rate refers to the proportion of the bandwidth actually used when the first image processing unit is working to the maximum bandwidth. For example, assuming that the maximum bandwidth of the first image processing unit is 100 / Mbps and the actually used bandwidth is 40 / Mbps, the bandwidth utilization rate is 40%, the IO rate refers to the speed of the first image processing unit when performing data reading and writing operations, the CPU load weight refers to the influence of the CPU utilization rate on the running load, the memory load weight refers to the influence of the memory usage on the running load, the bandwidth load weight refers to the influence of the bandwidth utilization rate on the running load, and the IO load weight refers to the influence of the IO rate on the running load.

[0146] S5. Based on the instruction analysis result and the image processing unit, perform image preprocessing on the collected image of the recycled steel to obtain a preprocessed image, wherein the image preprocessing includes: median filtering and binarization.

[0147] It can be explained that the pre-processed image refers to the image obtained after the recycled steel collection image is subjected to median filtering and binarization processing, which is the existing technology and will not be described in detail here.

[0148] S6. Perform fuzzy labeling on the preprocessed image using a preset fuzzy labeling method to obtain a fuzzy boundary set.

[0149] It can be explained that the fuzzy labeling method is eight-connected. Using the fuzzy labeling method to perform fuzzy labeling on the preprocessed image means using eight-connected to label the preprocessed image, obtaining several connected areas, and assigning a unique label to each connected area to distinguish the connected areas. The fuzzy boundary set refers to a set formed by several connected areas. Eight-connected is a prior art and will not be repeated here.

[0150] S7. Extract fuzzy boundaries in sequence from the fuzzy boundary set.

[0151] Interpretable, fuzzy boundaries refer to connected areas obtained after labeling using fuzzy labeling, and each connected area is a fuzzy boundary.

[0152] S8. Using a preset defect threshold, determine whether the fuzzy boundary is a preset defect boundary.

[0153] It can be explained that the defect threshold refers to the standard value for judging whether a fuzzy boundary is a defect boundary, and the defect boundary refers to a fuzzy boundary whose area is greater than or equal to the defect threshold.

[0154] In detail, the using a preset defect threshold to determine whether the fuzzy boundary is a defect boundary includes:

[0155] Acquire the total number of fuzzy boundary pixels of the fuzzy boundary and the pixel equivalent of the fuzzy boundary;

[0156] The fuzzy boundary area of ​​the fuzzy boundary is calculated based on the total number of fuzzy boundary pixels of the fuzzy boundary and the pixel equivalent of the fuzzy boundary:

[0157] M=S×Z

[0158] Among them, M refers to the fuzzy boundary area, S refers to the total number of fuzzy boundary pixels, and Z refers to the pixel equivalent;

[0159] Determining whether the fuzzy boundary area is smaller than the defect threshold;

[0160] If the fuzzy boundary area is smaller than the defect threshold, confirming that the fuzzy boundary is not a defect boundary;

[0161] If the fuzzy boundary area is not less than the defect threshold, the fuzzy boundary is confirmed to be a defect boundary.

[0162] To be explained, the total number of fuzzy boundary pixels refers to the total number of pixels in the fuzzy boundary, the pixel equivalent refers to the actual physical size represented by a pixel in the digital image, and the fuzzy boundary area refers to the area of ​​the fuzzy boundary.

[0163] If the fuzzy boundary is not a defect boundary, the process returns to the above step of sequentially extracting fuzzy boundaries from the fuzzy boundary set.

[0164] If the fuzzy boundary is a defect boundary, S9 is executed to determine the defect type of the defect boundary based on a preset defect standard, wherein the defect type includes: crack defect, hole defect and corrosion defect.

[0165] Explainable, defect type refers to various types of defect boundaries, including: cracks, holes and rust, crack defects refer to defects with crack type, hole defects refer to defects with hole type, and rust defects refer to defects with rust type.

[0166] In detail, the determining the defect type of the defect boundary based on a preset defect standard includes:

[0167] Obtaining a defect boundary area of ​​the defect boundary, and determining whether the defect boundary area is smaller than a preset crack area threshold;

[0168] If the defect boundary area is smaller than the crack area threshold, confirming that the defect type of the defect boundary is a crack defect;

[0169] If the defect boundary area is not less than the crack area threshold, then calculating the defect boundary roundness of the defect boundary;

[0170] The calculation method is as follows:

[0171] Obtain defective pixels, the total number of defective pixels, and the horizontal and vertical coordinates of the defective pixels at the defect boundary;

[0172] The perimeter of the defect boundary is calculated using the preset dist function, the defect pixels of the defect boundary, the total number of defect pixels, the abscissa and ordinate of the defect pixels:

[0173]

[0174] Among them, X refers to the perimeter of the defect boundary, E refers to the total number of defect pixels, dist refers to the dist function, and x τ Refers to the horizontal coordinate of the τth defective pixel, y τ Refers to the ordinate of the τth defective pixel, x τ+1 Refers to the horizontal coordinate of the τ+1th defective pixel, y τ+1 Refers to the ordinate of the τ+1th defective pixel;

[0175] Calculate the roundness of the defect boundary:

[0176]

[0177] Among them, Y refers to the roundness of the defect boundary, A refers to the area of ​​the defect boundary, X refers to the circumference of the defect boundary, and π refers to pi;

[0178] Determining whether the defect boundary roundness of the defect boundary is less than a preset hole roundness threshold;

[0179] If the defect boundary roundness of the defect boundary is not less than the hole roundness threshold, confirming that the defect type of the defect boundary is a hole defect;

[0180] If the defect boundary roundness of the defect boundary is less than the hole roundness threshold, it is confirmed that the defect type of the defect boundary is a corrosion defect.

[0181] It can be understood that the defect boundary area refers to the area of ​​the defect boundary, the crack area threshold refers to the value for judging whether the defect type of the defect boundary is a crack defect, when the defect boundary area is less than the crack area threshold, the defect type of the defect boundary is considered to be a crack defect, the defect boundary roundness refers to the regularity of the defect boundary shape, which is used to measure the degree to which the defect boundary is close to a circle, the total number of defect pixels refers to the total number of pixels within the defect boundary, the defect pixels refer to the pixels within the defect boundary, the defect boundary perimeter refers to the perimeter of the defect boundary, and the hole roundness threshold refers to the value for judging whether the defect type of the defect boundary is a hole defect.

[0182] S10. Calculate a defect score of the defect boundary based on the defect type.

[0183] Explainably, the defect score refers to the severity of the defect. The higher the defect score, the more serious the defect and the lower the recycling value.

[0184] In detail, the defect score of the defect boundary is calculated based on the defect type, including:

[0185] If the defect type is a crack defect, then obtaining the crack area, crack depth and crack length of the crack defect;

[0186] Substitute the crack area, crack depth and crack length into the preset crack area standard, crack depth standard and crack length standard respectively to obtain the crack area score, crack depth score and crack length score, wherein the crack area standard is: (0,5] cm 2 : 2 points, (5, 10] cm 2 :6 points, (10, +∞) cm 2 : 10 points, the crack depth standard is: (0, 0.5] mm: 2 points, (0.5, 1.5] mm: 6 points, (1.5, +∞) mm: 10 points, the crack length standard is: (0, 2] cm: 2 points, (2, 6] cm: 6 points, (6, +∞) cm: 10 points;

[0187] The defect score of the crack defect is calculated using the crack area score, crack depth score and crack length score:

[0188] U1=b1+b2+b3

[0189] Among them, U1 refers to the defect score of crack defect, b1 refers to the crack area score, b2 refers to the crack depth score, and b3 refers to the crack length score;

[0190] If the defect type is a hole defect, then obtaining the hole diameter, hole depth and hole roundness of the hole defect;

[0191] Substitute the hole diameter, hole depth and hole roundness into the preset hole diameter standard, hole depth standard and hole roundness standard to obtain the hole diameter score, hole depth score and hole roundness score, wherein the hole diameter standard is: (0, 0.5] cm: 2 points, (0.5, 4] cm: 6 points, (4, +∞) cm: 10 points, the hole depth standard is: (0, 0.5] mm: 2 points, (0.5, 3] mm: 6 points, (3, +∞) mm: 10 points, the hole roundness standard is: (0, 0.5]: 10 points, (0.5, 0.8], 6 points, (0.8, 1.0): 2 points;

[0192] The defect score of the hole defect type is calculated using the hole diameter score, hole depth score and hole roundness score:

[0193] U2=g1+g2+g3

[0194] Among them, U2 refers to the defect score of hole defect, g1 refers to the hole diameter score, g2 refers to the hole depth score, and g3 refers to the hole roundness score;

[0195] If the defect type is a rust defect, the rust area, rust depth and texture complexity of the rust defect are obtained, wherein the texture complexity includes: smooth, rough and extremely rough;

[0196] Substitute the rust area, rust depth, and texture complexity into the preset rust area standard, rust depth standard, and texture complexity standard to obtain the rust area score, rust depth score, and texture complexity score, wherein the rust area standard is: (0, 5] cm 2 : 2 points, (5, 10] cm 2 :6 points, (10, +∞) cm 2 : 10 points, the rust depth standard is: (0, 1] mm: 2 points, (1, 4] mm: 6 points, (4, +∞) mm: 10 points, the texture complexity standard is: smooth: 2 points, rough: 6 points, extremely rough: 10 points;

[0197] The defect score of the rust defect type is calculated using the rust area score, rust depth score and texture complexity score:

[0198] U3=k1+k2+k3

[0199] Among them, U3 refers to the defect score of rust defect, k1 refers to the rust area score, k2 refers to the rust depth score, and k3 refers to the texture complexity score.

[0200] It can be understood that crack area refers to the area of ​​the crack defect, crack depth refers to the depth of the crack defect, crack length refers to the length of the crack defect, crack area standard refers to the standard for scoring the crack area, crack depth standard refers to the standard for scoring the crack depth, crack length standard refers to the standard for scoring the crack length, the crack area score refers to the score obtained by scoring the crack area using the crack area standard, the crack depth score refers to the score obtained by scoring the crack depth using the crack depth standard, and the crack length score refers to the score obtained by scoring the crack length using the crack length standard.

[0201] It can be explained that the hole diameter refers to the diameter of the hole defect, the hole depth refers to the depth of the hole defect, the hole roundness refers to the roundness of the hole defect, the hole diameter standard refers to the standard for scoring the hole diameter, the hole depth standard refers to the standard for scoring the hole depth, the hole roundness standard refers to the standard for scoring the hole roundness, the hole diameter score refers to the score obtained by scoring the hole diameter using the hole diameter standard, the hole depth score refers to the score obtained by scoring the hole depth using the hole depth standard, and the hole roundness score refers to the score obtained by scoring the hole roundness using the hole roundness standard.

[0202] It can be understood that the rust area refers to the area of ​​the rust defect, the rust depth refers to the depth of the rust defect, the texture complexity refers to the roughness of the surface texture of the rust defect, the rust area standard refers to the standard for scoring the rust area, the rust depth standard refers to the standard for scoring the rust depth, the texture complexity standard refers to the standard for scoring the texture complexity, the rust area score refers to the score obtained by scoring the rust area using the rust area standard, the rust depth score refers to the score obtained by scoring the rust area using the rust depth standard, and the texture complexity score refers to the score obtained by scoring the texture complexity using the texture complexity standard.

[0203] S11, judging whether the fuzzy boundary is extracted.

[0204] If the fuzzy boundary extraction is not completed, return to the above step of sequentially extracting fuzzy boundaries from the fuzzy boundary set.

[0205] If the fuzzy boundary extraction is completed, S12 is executed to summarize the defect scores to obtain a defect score set, and a comprehensive defect score is calculated based on the defect score set.

[0206] Interpretably, the defect score set refers to the set formed by the defect scores of all defect boundaries, and the comprehensive defect score refers to the score calculated based on the defect scores of all defect boundaries, which is used to comprehensively evaluate scrap recycled steel.

[0207] In detail, the calculating of the comprehensive defect score based on the defect score set includes:

[0208] Dividing the defect score set into a crack score set, a hole score set and a corrosion score set based on the defect type, and obtaining the number of crack scores, the number of hole scores, the number of corrosion scores, the crack scores, the hole scores and the corrosion scores by using the crack score set, the hole score set and the corrosion score set;

[0209] Obtain crack score weights, hole score weights, and corrosion score weights, and calculate a comprehensive defect score based on the number of crack scores, the number of hole scores, the number of corrosion scores, the crack score, the hole score, the corrosion score, the crack score weights, the hole score weights, and the corrosion score weights:

[0210]

[0211] Among them, U4 refers to the comprehensive defect score, l1 refers to the crack score weight, f refers to the number of crack scores, and h α refers to the αth crack score, l2 refers to the hole score weight, t refers to the number of hole scores, c β refers to the βth hole score, l3 refers to the rust score weight, o refers to the number of rust scores, j γ Refers to the γth rust score.

[0212] Explainable, crack score set refers to the set formed by the defect scores of all defects whose defect types are crack defects in the defect score set, hole score set refers to the set formed by the defect scores of all defects whose defect types are hole defects in the defect score set, corrosion score set refers to the set formed by the defect scores of all defects whose defect types are corrosion defects in the defect score set, the number of crack scores refers to the number of crack scores in the crack score set, crack scores refer to the defect scores of defects whose defect types are crack defects, the number of hole scores refers to the number of hole scores in the hole score set, hole scores refer to the defect scores of defects whose defect types are hole defects, the number of corrosion scores refers to the number of rust scores in the rust score set, rust scores refer to the defect scores of defects whose defect types are rust defects, the crack score weight refers to the influence of the rust score on the comprehensive defect score, the hole score weight refers to the influence of the hole score on the comprehensive defect score, and the corrosion score weight refers to the influence of the rust score on the comprehensive defect score.

[0213] S13. Evaluate and grade the scrap recycled steel based on the preset scoring level and the comprehensive defect score to obtain graded recycled steel.

[0214] It can be explained that the scoring levels are: (0, 3] points: excellent grade, (3, 6] points: good grade, (6, 10] points: damaged grade. The comprehensive defect score is substituted into the corresponding scoring level to obtain graded recycled steel. The excellent grade of scrap recycled steel has no obvious structural defects and can continue to be used. The good grade of scrap recycled steel has slight defects and can be used normally after recycling and repair. The damaged grade of scrap recycled steel has serious defects and its structural integrity is damaged, so it cannot be used.

[0215] In order to solve the problems described in the background technology, the present invention realizes the automatic detection and evaluation of the scrap recycled steel by combining the scoring level and the comprehensive defect score to perform multi-step processing on the image of the scrap recycled steel; firstly, the image acquisition unit is used to perform automatic image acquisition on the scrap recycled steel. The automatic image acquisition can quickly and continuously obtain the image of the scrap recycled steel, which reduces the waste of manpower and ensures that the real state of the surface of the scrap recycled steel can be captured in time, which provides accurate image information for subsequent processing and improves the efficiency of overall detection and evaluation; secondly, the collected image of the recycled steel is preprocessed by median filtering and binarization. The median filtering and binarization processing can effectively remove background noise and highlight important features, making defects easier to observe in the image, and making fuzzy marking and defect recognition more efficient, thereby improving the accuracy and efficiency of detection; thereafter, the preprocessed image is fuzzy marked by using the fuzzy marking method to obtain a fuzzy boundary set. The fuzzy boundary set can help to more accurately define the defect area, thereby improving the efficiency and accuracy of defect recognition; further, the fuzzy boundary sets are extracted in sequence in the fuzzy boundary set. Boundary, and use the defect threshold to determine whether the extracted fuzzy boundary is a defect boundary. When it is a defect boundary, the defect type of the defect boundary is determined based on the defect standard. Determining the defect type of the defect boundary can reduce subjective judgment and improve the reliability of subsequent defect score calculation; then, the defect score of the defect boundary is calculated based on the defect type, and each defect is scored to quantify the severity of the defect, so that the evaluation result is more intuitive, which not only provides a data basis for evaluation and grading, but also reduces the error of human judgment; then, when the fuzzy boundary is extracted, the defect score is summarized to obtain a defect score set, and a comprehensive defect score is calculated based on the defect score set. The comprehensive defect score is calculated based on the defect score, which can make the quality status of the scrap recycled steel clearer and improve the evaluation efficiency; finally, the scrap recycled steel is evaluated and graded based on the scoring level and the comprehensive defect score, and effective grading is performed based on the comprehensive defect score and the scoring level, which can quickly evaluate the quality of the scrap recycled steel. This standardized evaluation process simplifies the operation process and makes the evaluation of the scrap recycled steel more efficient. Therefore, the present invention can improve the detection and evaluation efficiency of scrap recycled steel.

[0216] like Figure 2, which is a functional module diagram of an automated detection and evaluation system for scrap recycled steel provided in one embodiment of the present invention.

[0217] The automated detection and evaluation system 100 for scrap recycled steel of the present invention can be installed in an electronic device. According to the functions to be implemented, the automated detection and evaluation system 100 for scrap recycled steel can include an image acquisition module 101, an image preprocessing module 102, a defect boundary judgment module 103 and a recycled steel classification module 104. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0218] The image acquisition module 101 is used to receive a detection instruction and start a pre-built image detection unit based on the detection instruction, wherein the image detection unit includes: an image acquisition unit, an image storage unit and an image processing unit, and the image acquisition unit is used to perform automatic image acquisition on the preset scrap recycled steel to obtain a recycled steel acquisition image and generate an image storage instruction;

[0219] The image preprocessing module 102 is used to receive the image storage instruction by using the image storage unit, store the recycled steel collected image in the image storage unit and generate an image processing instruction, receive the image processing instruction by using the image processing unit and perform processing instruction analysis to obtain an instruction analysis result, and perform image preprocessing on the recycled steel collected image based on the instruction analysis result and the image processing unit to obtain a preprocessed image, wherein the image preprocessing includes: median filtering and binarization;

[0220] The defect boundary judgment module 103 is used to perform fuzzy marking on the pre-processed image using a preset fuzzy marking method to obtain a fuzzy boundary set, extract fuzzy boundaries in the fuzzy boundary set in sequence, and use a preset defect threshold to determine whether the fuzzy boundary is a preset defect boundary. If the fuzzy boundary is not a defect boundary, return to the above step of sequentially extracting fuzzy boundaries in the fuzzy boundary set. If the fuzzy boundary is a defect boundary, determine the defect type of the defect boundary based on a preset defect standard, wherein the defect types include: crack defects, hole defects and corrosion defects.

[0221] The recycled steel grading module 104 is used to calculate the defect score of the defect boundary based on the defect type, determine whether the fuzzy boundary extraction has been completed, and if the fuzzy boundary extraction has not been completed, return to the above step of extracting the fuzzy boundaries in the fuzzy boundary set in sequence; if the fuzzy boundary extraction has been completed, summarize the defect scores to obtain a defect score set, and calculate a comprehensive defect score based on the defect score set, and evaluate and grade the scrap recycled steel based on the preset scoring level and the comprehensive defect score.

[0222] In detail, the modules in the automated detection and evaluation system 100 for recycled steel in the embodiment of the present invention are used in the same manner as described above. Figure 1 The same technical means are used as the automated detection and evaluation method for scrap recycled steel described in , and can produce the same technical effects, so they will not be repeated here.

[0223] like Figure 3 , which is a schematic diagram of the structure of an electronic device for realizing an automated detection and evaluation method for scrap recycled steel provided by an embodiment of the present invention.

[0224] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as an automated detection and evaluation method program for scrap recycled steel.

[0225] Wherein, the memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (for example: SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. The memory 11 may be an internal storage unit of the electronic device 1 in some embodiments, such as a mobile hard disk of the electronic device 1. The memory 11 may also be an external storage device of the electronic device 1 in other embodiments, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Further, the memory 11 also includes an internal storage unit of the electronic device 1 and an external storage device. The memory 11 can not only be used to store application software and various types of data installed in the electronic device 1, such as the code of the program of the automated detection and evaluation method of scrap recycled steel, but also can be used to temporarily store data that has been output or is to be output.

[0226] The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes or executes programs or modules stored in the memory 11 (such as the automated detection and evaluation method program for scrap recycled steel, etc.), and calls the data stored in the memory 11 to execute various functions of the electronic device 1 and process data.

[0227] The bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0228] Figure 3 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0229] For example, although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to each component. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include any components such as one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, etc. The electronic device 1 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0230] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0231] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0232] The program of the automated detection and evaluation method of scrap recycled steel stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve:

[0233] Receiving a detection instruction, and starting a pre-built image detection unit based on the detection instruction, wherein the image detection unit includes: an image acquisition unit, an image storage unit and an image processing unit;

[0234] Utilizing the image acquisition unit to perform automated image acquisition on the preset scrap recycled steel, obtaining the recycled steel acquisition image and generating an image storage instruction;

[0235] Using the image storage unit to receive the image storage instruction, storing the recycled steel collected image into the image storage unit and generating an image processing instruction;

[0236] Utilizing the image processing unit to receive the image processing instruction and perform processing instruction parsing to obtain an instruction parsing result;

[0237] Based on the instruction analysis result and the image processing unit, image preprocessing is performed on the collected image of the recycled steel to obtain a preprocessed image, wherein the image preprocessing includes: median filtering and binarization;

[0238] Performing fuzzy labeling on the preprocessed image using a preset fuzzy labeling method to obtain a fuzzy boundary set;

[0239] Extracting fuzzy boundaries in sequence from the fuzzy boundary set;

[0240] Using a preset defect threshold to determine whether the fuzzy boundary is a preset defect boundary;

[0241] If the fuzzy boundary is not a defect boundary, returning to the above step of sequentially extracting fuzzy boundaries from the fuzzy boundary set;

[0242] If the fuzzy boundary is a defect boundary, the defect type of the defect boundary is determined based on a preset defect standard, wherein the defect type includes: crack defect, hole defect and corrosion defect;

[0243] Calculating a defect score for the defect boundary based on the defect type;

[0244] Determining whether the fuzzy boundary is extracted;

[0245] If the fuzzy boundary extraction is not completed, returning to the above step of sequentially extracting fuzzy boundaries in the fuzzy boundary set;

[0246] If the fuzzy boundary extraction is completed, the defect scores are summarized to obtain a defect score set, and a comprehensive defect score is calculated based on the defect score set;

[0247] The scrap recycled steel is evaluated and graded based on a preset scoring level and the comprehensive defect score to obtain graded recycled steel.

[0248] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figures 1 to 3 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0249] Furthermore, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0250] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement:

[0251] Receiving a detection instruction, and starting a pre-built image detection unit based on the detection instruction, wherein the image detection unit includes: an image acquisition unit, an image storage unit and an image processing unit;

[0252] Utilizing the image acquisition unit to perform automated image acquisition on the preset scrap recycled steel, obtaining the recycled steel acquisition image and generating an image storage instruction;

[0253] Using the image storage unit to receive the image storage instruction, storing the recycled steel collected image into the image storage unit and generating an image processing instruction;

[0254] Utilizing the image processing unit to receive the image processing instruction and perform processing instruction parsing to obtain an instruction parsing result;

[0255] Based on the instruction analysis result and the image processing unit, image preprocessing is performed on the collected image of the recycled steel to obtain a preprocessed image, wherein the image preprocessing includes: median filtering and binarization;

[0256] Performing fuzzy labeling on the preprocessed image using a preset fuzzy labeling method to obtain a fuzzy boundary set;

[0257] Extracting fuzzy boundaries in sequence from the fuzzy boundary set;

[0258] Using a preset defect threshold to determine whether the fuzzy boundary is a preset defect boundary;

[0259] If the fuzzy boundary is not a defect boundary, returning to the above step of sequentially extracting fuzzy boundaries from the fuzzy boundary set;

[0260] If the fuzzy boundary is a defect boundary, the defect type of the defect boundary is determined based on a preset defect standard, wherein the defect type includes: crack defect, hole defect and corrosion defect;

[0261] Calculating a defect score for the defect boundary based on the defect type;

[0262] Determining whether the fuzzy boundary is extracted;

[0263] If the fuzzy boundary extraction is not completed, returning to the above step of sequentially extracting fuzzy boundaries in the fuzzy boundary set;

[0264] If the fuzzy boundary extraction is completed, the defect scores are summarized to obtain a defect score set, and a comprehensive defect score is calculated based on the defect score set;

[0265] The scrap recycled steel is evaluated and graded based on a preset scoring level and the comprehensive defect score to obtain graded recycled steel.

[0266] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, and actual implementation may have other division methods.

[0267] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0268] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0269] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0270] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. An automated detection and evaluation method for scrap recycled steel, characterized in that: The method comprises: Receiving a detection instruction, and starting a pre-built image detection unit based on the detection instruction, wherein the image detection unit includes: an image acquisition unit, an image storage unit and an image processing unit; Utilizing the image acquisition unit to perform automated image acquisition on the preset scrap recycled steel, obtaining the recycled steel acquisition image and generating an image storage instruction; Using the image storage unit to receive the image storage instruction, storing the recycled steel collected image into the image storage unit and generating an image processing instruction; Utilizing the image processing unit to receive the image processing instruction and perform processing instruction parsing to obtain an instruction parsing result; Based on the instruction analysis result and the image processing unit, image preprocessing is performed on the collected image of the recycled steel to obtain a preprocessed image, wherein the image preprocessing includes: median filtering and binarization; Performing fuzzy labeling on the preprocessed image using a preset fuzzy labeling method to obtain a fuzzy boundary set; Extracting fuzzy boundaries in sequence from the fuzzy boundary set; Using a preset defect threshold to determine whether the fuzzy boundary is a preset defect boundary; If the fuzzy boundary is not a defect boundary, returning to the above step of sequentially extracting fuzzy boundaries from the fuzzy boundary set; If the fuzzy boundary is a defect boundary, the defect type of the defect boundary is determined based on a preset defect standard, wherein the defect type includes: crack defect, hole defect and corrosion defect; Calculating a defect score for the defect boundary based on the defect type; Determining whether the fuzzy boundary is extracted; If the fuzzy boundary extraction is not completed, returning to the above step of sequentially extracting fuzzy boundaries in the fuzzy boundary set; If the fuzzy boundary extraction is completed, the defect scores are summarized to obtain a defect score set, and a comprehensive defect score is calculated based on the defect score set; The scrap recycled steel is evaluated and graded based on a preset scoring level and the comprehensive defect score to obtain graded recycled steel.

2. The automated detection and evaluation method for recycled scrap steel according to claim 1, characterized in that: The method of using the image acquisition unit to perform automatic image acquisition on the preset scrap recycled steel to obtain the recycled steel acquisition image and generate an image storage instruction includes: Utilizing the image acquisition unit to continuously photograph the scrap recycled steel for 3 seconds at a shooting frequency of 5 images / s to obtain a recycled steel image set; Based on the recycled steel image set, two images with the highest definition are selected to obtain a first representative image and a second representative image; Acquire a first pixel value set of the first representative image and a second pixel value set of the second representative image; Expanding the first pixel value set and the second pixel value set into a one-dimensional array to obtain a first pixel array and a second pixel array; Calculating the image similarity between the first representative image and the second representative image based on the first pixel array and the second pixel array; Determining whether the image similarity is greater than a preset similarity threshold; If the image similarity is greater than a similarity threshold, randomly selecting one of the first representative image and the second representative image as a recycled steel collection image, and generating an image storage instruction based on the recycled steel collection image; If the image similarity is not greater than the similarity threshold, the first representative image and the second representative image are used as recycled steel collection images, and an image storage instruction is generated based on the recycled steel collection images.

3. The automated detection and evaluation method for recycled scrap steel according to claim 2, characterized in that: The calculating the image similarity between the first representative image and the second representative image based on the first pixel array and the second pixel array includes: Acquire a total number of pixels based on the first pixel array and the second pixel array, wherein the total number of pixels is the total number of pixels of the first pixel array or the total number of pixels of the second pixel array, and the total number of pixels of the first pixel array is equal to the total number of pixels of the second pixel array; The image similarity between the first representative image and the second representative image is calculated using the total number of pixels, the first pixel array, and the second pixel array: Among them, T refers to the image similarity, n refers to the total number of pixels, and Q i Refers to the i-th pixel value in the first pixel array, W i Refers to the i-th pixel value in the second pixel array.

4. The automated detection and evaluation method for recycled scrap steel according to claim 1, characterized in that: The step of using the image processing unit to receive the image processing instruction and perform processing instruction parsing to obtain an instruction parsing result includes: The image processing unit includes: a first image processing unit and a second image processing unit; Processing instruction parsing is as follows: Calculating the operating load of the first image processing unit, and determining whether the operating load is less than a preset operating load threshold; If the operating load is less than the operating load threshold, determining the instruction analysis result as inputting the recycled steel acquisition image into the first image processing unit; If the operating load is equal to or greater than the operating load threshold, the instruction analysis result is determined to input the recycled steel acquisition image into the second image processing unit, and after the operating load is less than the operating load threshold, the recycled steel acquisition image is input into the first image acquisition unit.

5. The automated detection and evaluation method for recycled scrap steel according to claim 4, characterized in that: The calculating the operation load of the first image processing unit includes: Obtaining the CPU usage, memory usage, bandwidth usage and IO rate of the first image processing unit; Obtaining a CPU load weight, a memory load weight, a bandwidth load weight, and an IO load weight of the first image processing unit; The operation load of the first image processing unit is calculated based on the CPU usage rate, memory usage, bandwidth usage rate, IO rate, CPU load weight, memory load weight, bandwidth load weight and IO load weight: H=ρ1×R1+ρ2×R2+ρ3×R3+ρ4×R4 Among them, H refers to the operating load of the first image processing unit, ρ1 refers to the CPU load weight, R1 refers to the CPU usage rate, ρ2 refers to the memory load weight, R2 refers to the memory usage, ρ3 refers to the bandwidth load weight, R3 refers to the bandwidth usage rate, ρ4 refers to the IO load weight, and R4 refers to the IO rate.

6. The automated detection and evaluation method for recycled scrap steel according to claim 1, characterized in that: The using a preset defect threshold to determine whether the fuzzy boundary is a defect boundary includes: Acquire the total number of fuzzy boundary pixels of the fuzzy boundary and the pixel equivalent of the fuzzy boundary; The fuzzy boundary area of ​​the fuzzy boundary is calculated based on the total number of fuzzy boundary pixels of the fuzzy boundary and the pixel equivalent of the fuzzy boundary: M=S×Z Among them, M refers to the fuzzy boundary area, S refers to the total number of fuzzy boundary pixels, and Z refers to the pixel equivalent; Determining whether the fuzzy boundary area is smaller than the defect threshold; If the fuzzy boundary area is smaller than the defect threshold, confirming that the fuzzy boundary is not a defect boundary; If the fuzzy boundary area is not less than the defect threshold, the fuzzy boundary is confirmed to be a defect boundary.

7. The automated detection and evaluation method for recycled scrap steel according to claim 1, characterized in that: The determining the defect type of the defect boundary based on a preset defect standard includes: Obtaining a defect boundary area of ​​the defect boundary, and determining whether the defect boundary area is smaller than a preset crack area threshold; If the defect boundary area is smaller than the crack area threshold, confirming that the defect type of the defect boundary is a crack defect; If the defect boundary area is not less than the crack area threshold, then calculating the defect boundary roundness of the defect boundary; The calculation method is as follows: Obtain defective pixels, the total number of defective pixels, and the horizontal and vertical coordinates of the defective pixels at the defect boundary; The perimeter of the defect boundary is calculated using the preset dist function, the defect pixels of the defect boundary, the total number of defect pixels, the abscissa and ordinate of the defect pixels: Among them, X refers to the perimeter of the defect boundary, E refers to the total number of defect pixels, dist refers to the dist function, and x τ Refers to the horizontal coordinate of the τth defective pixel, y τ Refers to the ordinate of the τth defective pixel, x τ+1 Refers to the horizontal coordinate of the τ+1th defective pixel, y τ+1 Refers to the ordinate of the τ+1th defective pixel; Calculate the roundness of the defect boundary: Among them, Y refers to the roundness of the defect boundary, A refers to the area of ​​the defect boundary, X refers to the circumference of the defect boundary, and π refers to pi; Determining whether the defect boundary roundness of the defect boundary is less than a preset hole roundness threshold; If the defect boundary roundness of the defect boundary is not less than the hole roundness threshold, confirming that the defect type of the defect boundary is a hole defect; If the defect boundary roundness of the defect boundary is less than the hole roundness threshold, it is confirmed that the defect type of the defect boundary is a corrosion defect.

8. The automated detection and evaluation method for recycled scrap steel according to claim 1, characterized in that: The calculating the defect score of the defect boundary based on the defect type includes: If the defect type is a crack defect, then obtaining the crack area, crack depth and crack length of the crack defect; Substitute the crack area, crack depth and crack length into the preset crack area standard, crack depth standard and crack length standard respectively to obtain the crack area score, crack depth score and crack length score, wherein the crack area standard is: (0,5] cm 2 : 2 points, (5, 10] cm 2 :6 points, (10, +∞)cm 2 : 10 points, the crack depth standard is: (0, 0.5] mm: 2 points, (0.5, 1.5] mm: 6 points, (1.5, +∞) mm: 10 points, the crack length standard is: (0, 2] cm: 2 points, (2, 6] cm: 6 points, (6, +∞) cm: 10 points; The defect score of the crack defect is calculated using the crack area score, crack depth score and crack length score: U1=b1+b2+b3 Among them, U1 refers to the defect score of crack defect, b1 refers to the crack area score, b2 refers to the crack depth score, and b3 refers to the crack length score; If the defect type is a hole defect, then obtaining the hole diameter, hole depth and hole roundness of the hole defect; Substitute the hole diameter, hole depth and hole roundness into the preset hole diameter standard, hole depth standard and hole roundness standard to obtain the hole diameter score, hole depth score and hole roundness score, wherein the hole diameter standard is: (0, 0.5] cm: 2 points, (0.5, 4] cm: 6 points, (4, +∞) cm: 10 points, the hole depth standard is: (0, 0.5] mm: 2 points, (0.5, 3] mm: 6 points, (3, +∞) mm: 10 points, the hole roundness standard is: (0, 0.5]: 10 points, (0.5, 0.8], 6 points, (0.8, 1.0): 2 points; The defect score of the hole defect type is calculated using the hole diameter score, hole depth score and hole roundness score: U2=g1+g2+g3 Among them, U2 refers to the defect score of hole defect, g1 refers to the hole diameter score, g2 refers to the hole depth score, and g3 refers to the hole roundness score; If the defect type is a rust defect, the rust area, rust depth and texture complexity of the rust defect are obtained, wherein the texture complexity includes: smooth, rough and extremely rough; Substitute the rust area, rust depth, and texture complexity into the preset rust area standard, rust depth standard, and texture complexity standard to obtain the rust area score, rust depth score, and texture complexity score, wherein the rust area standard is: (0, 5] cm 2 : 2 points, (5, 10] cm 2 :6 points, (10, +∞)cm 2 : 10 points, the rust depth standard is: (0, 1] mm: 2 points, (1, 4] mm: 6 points, (4, +∞) mm: 10 points, the texture complexity standard is: smooth: 2 points, rough: 6 points, extremely rough: 10 points; The defect score of the rust defect type is calculated using the rust area score, rust depth score and texture complexity score: U3=k1+k2+k3 Among them, U3 refers to the defect score of rust defect, k1 refers to the rust area score, k2 refers to the rust depth score, and k3 refers to the texture complexity score.

9. The automated detection and evaluation method for recycled scrap steel according to claim 1, characterized in that: The calculating a comprehensive defect score based on the defect score set includes: Dividing the defect score set into a crack score set, a hole score set and a corrosion score set based on the defect type, and obtaining the number of crack scores, the number of hole scores, the number of corrosion scores, the crack scores, the hole scores and the corrosion scores by using the crack score set, the hole score set and the corrosion score set; Obtain crack score weights, hole score weights, and corrosion score weights, and calculate a comprehensive defect score based on the number of crack scores, the number of hole scores, the number of corrosion scores, the crack score, the hole score, the corrosion score, the crack score weights, the hole score weights, and the corrosion score weights: Among them, U4 refers to the comprehensive defect score, l1 refers to the crack score weight, f refers to the number of crack scores, and h α refers to the αth crack score, l2 refers to the hole score weight, t refers to the number of hole scores, c β refers to the βth hole score, l3 refers to the rust score weight, o refers to the number of rust scores, j γ Refers to the γth rust score.

10. An automated detection and evaluation system for recycled scrap steel, characterized in that: The system comprises: An image acquisition module is used to receive a detection instruction and start a pre-built image detection unit based on the detection instruction, wherein the image detection unit includes: an image acquisition unit, an image storage unit and an image processing unit, and the image acquisition unit is used to perform automatic image acquisition on the preset scrap recycled steel to obtain a recycled steel acquisition image and generate an image storage instruction; An image preprocessing module is used to receive the image storage instruction by using the image storage unit, store the recycled steel collected image in the image storage unit and generate an image processing instruction, receive the image processing instruction by using the image processing unit and perform processing instruction analysis to obtain an instruction analysis result, and perform image preprocessing on the recycled steel collected image based on the instruction analysis result and the image processing unit to obtain a preprocessed image, wherein the image preprocessing includes: median filtering and binarization; a defect boundary judgment module, configured to perform fuzzy marking on the preprocessed image using a preset fuzzy marking method to obtain a fuzzy boundary set, extract fuzzy boundaries in the fuzzy boundary set in sequence, and determine whether the fuzzy boundary is a preset defect boundary using a preset defect threshold; if the fuzzy boundary is not a defect boundary, return to the above step of sequentially extracting fuzzy boundaries in the fuzzy boundary set; if the fuzzy boundary is a defect boundary, determine the defect type of the defect boundary based on a preset defect standard, wherein the defect types include: crack defects, hole defects, and corrosion defects; The recycled steel grading module is used to calculate the defect score of the defect boundary based on the defect type, determine whether the fuzzy boundary extraction has been completed, and if the fuzzy boundary extraction has not been completed, return to the above step of extracting fuzzy boundaries in the fuzzy boundary set in sequence; if the fuzzy boundary extraction is completed, summarize the defect scores to obtain a defect score set, and calculate a comprehensive defect score based on the defect score set, and evaluate and grade the scrap recycled steel based on a preset scoring level and the comprehensive defect score.

Images