Multi-goods-type material information identification processing method and system

Abstract

The invention relates to the technical field of image recognition, in particular to a multi-goods-type material information recognition processing method and system. The method comprises the following steps: acquiring a material image in a Lab mode through image acquisition equipment; constructing a first image generation model, and combining the material image and the first image generation model to obtain a first image of the material image; in the first image, extracting texture features of the material image; on the basis of various machine classification models, the credibility of the various machine classification models is obtained by utilizing texture features of the multi-goods-type material image test samples; and according to the credibility and the material image, obtaining the joint credibility of the machine classification model for identifying the multiple kinds of materials, and completing identification processing of the information of the multiple kinds of materials based on the joint credibility. According to the method and the system, the efficiency of classifying, sorting and grading various materials in the dry bulk cargo transfer and storage process is improved by accurately identifying the information of the various kinds of materials.

Description

Technical Field

[0001] The present invention relates to the field of image recognition technology, and in particular to a method and system for identifying and processing information of multiple types of materials. Background Art

[0002] With the acceleration of globalization and the continuous increase in trade, the construction and development of ports are directly related to the smoothness of international trade. Therefore, more and more countries and regions have begun to pay attention to the construction and development of ports. In my country's port construction, the construction of dry bulk ports has occupied an increasingly high proportion in recent years. In dry bulk ports, the goods unloaded from the ship are transferred to the yard through conveyor belts. However, due to the wide variety of dry bulk cargoes, automated, fast and accurate cargo grading and classification can greatly improve the port's operating efficiency. For the grading and classification of multiple types of goods, the primary task is to identify and process the information of multiple types of materials. Therefore, fast and accurate information identification and processing of multiple types of materials is a problem that needs to be solved urgently. Summary of the invention

[0003] In view of the shortcomings of the existing methods and the needs of practical applications, in order to improve the grading and classification capabilities of multiple types of materials and solve the problem of quickly and accurately identifying and processing the information of multiple types of materials. On the one hand, the present invention provides a method for identifying and processing information of multiple types of materials, comprising the following steps: Through the image acquisition device, the material image of the Lab mode is acquired; the first image generation model is constructed, and the first image of the material image is obtained by combining the material image and the first image generation model; in the first image, the texture features of the material image are extracted; based on multiple machine classification models, the texture features of the image test samples of multiple types of materials are used to obtain the credibility of multiple machine classification models; according to the credibility and the material image, the joint credibility of the machine classification model in identifying multiple types of materials is obtained, and the recognition processing of the information of multiple types of materials is completed based on the joint credibility. The present invention obtains the image data of multiple types of materials through machine vision, optimizes the image data and then extracts the texture features, and finally completes the recognition processing of the information of multiple types of materials based on the DS theory combined with the recognition results of multiple machine models, thereby improving the grading and classification capabilities of multiple types of materials and solving the problem of rapid and accurate information recognition and processing of multiple types of materials.

[0004] Optionally, the constructed first image generation model satisfies the following formula: ,in, Represents the generated image Pixels Channel value, Represents multiple types of material images Channel average, express Channel value adjustment factor, Represents the image of multiple types of materials Pixels Channel value standard deviation, Represents the generated image Pixels Channel value, Represents multiple types of material images Channel average, express Channel value adjustment factor, Represents the image of multiple types of materials Pixels Channel value standard deviation, Represents the generated image Pixels Channel value, Represents multiple types of material images Channel average, express Channel value adjustment factor, Represents the image of multiple types of materials Pixels Channel value standard deviation. The present invention corrects the image through the channel value adjustment factor, highlights the region of interest of the image, reduces the interference of the background image, is conducive to the extraction of features in the subsequent steps, and further ensures the recognition accuracy of the present invention.

[0005] Optionally, extracting the texture features of the material image in the first image comprises the following steps: Acquire representative pixel values ​​of the first image; obtain first coding values ​​and second coding values ​​of each pixel in the first image through a first coding algorithm, a second coding algorithm and the representative pixel values; and obtain texture features of the multiple types of materials by counting the histogram vectors of the coding values ​​of the first image.

[0006] Optionally, the obtaining of the representative pixel value of the first image satisfies the following formula: in, Indicates The representative pixel value of a pixel point is Represents the generated image Pixels Channel value, express Channel value adjustment factor, Represents the generated image Pixels Channel value, express Channel value adjustment factor, Represents the generated image Pixels Channel value, express Channel value adjustment factor: The representative pixel value provided by the present invention is combined with the channel value adjustment factor, and the obtained representative pixel value can reflect the color and brightness information of the image, which is further conducive to improving the ability of the present invention to extract image features.

[0007] Optionally, obtaining the first coding value by using the first coding algorithm comprises the following steps: Establish a nine-square grid with the pixel point as the center point; obtain the representative pixel value difference between the pixel point and other pixel points in the nine-square grid; mark the representative pixel value difference greater than the average value of the representative pixel value difference as 1, and mark the representative pixel value difference less than or equal to the average value of the representative pixel value difference as 0; Direction statistics mark, obtaining the first encoding value; Obtaining the second encoding value by using the second encoding algorithm includes the following steps: Establish a twenty-five-square grid with the pixel point as the center point; obtain 0 in the twenty-five-square grid, , , , , , , The first encoding value and the second encoding value obtained by the first encoding algorithm and the second encoding algorithm provided by the present invention can fully represent the image features and further improve the recognition accuracy of the present invention.

[0008] Optionally, the machine classification model includes a BP neural network model, a K-means classification model, a SVM classification model, a random forest model, a decision tree model and an integrated learning model. The present invention utilizes a variety of machine classification models to identify material information and verify each other based on the results, which is conducive to improving the recognition accuracy of the present invention.

[0009] Optionally, the credibility of the multiple machine classification models is obtained by using texture features of multiple material image test samples based on multiple machine classification models, satisfying the following formula: ,in, Indicates The credibility of the machine classification model, represents the number of machine classification models, Indicates the quantity of material type. Indicates Machine classification model recognition The success rate of the material Indicates Machine classification model recognition The reliability of the machine classification model provided by the present invention is evaluated for each machine classification model, which further ensures the accuracy of the material recognition of the present invention.

[0010] Optionally, the joint credibility of the machine classification model in identifying multiple types of materials is obtained based on the credibility and the material image, satisfying the following formula: in, Indicates Machine classification model recognition The combined reliability of the materials, Indicates Machine classification model recognition The success rate of the material Indicates The credibility of the machine classification model, Indicates the uniformity of machine classification model recognition. If the machine classification model recognizes the same material image, then If the machine classification model recognizes the material image differently, , Indicates the number of machine classification models. The joint credibility set in the present invention combines the uniformity of machine classification model recognition, which is conducive to improving the accuracy of the recognition probability of the material obtained by fusion through the DS evidence theory fusion system.

[0011] Optionally, completing the identification process of multiple types of material information based on the joint credibility includes the following steps: The DS evidence theory fusion system is used to fuse the joint credibility to obtain the recognition probability of the material; the recognition probability judgment rule is set, and the recognition probability and the recognition probability judgment rule are combined to complete the recognition processing of multi-type material information. The present invention determines the recognition result simply and quickly through the recognition probability judgment rule, thereby improving the recognition speed of the present invention.

[0012] In the second aspect, in order to efficiently execute the multi-type material information identification and processing method provided by the present invention, the present invention also provides a multi-type material information identification and processing system, including a processor, an input device, an output device and a memory, wherein the processor, the input device, the output device and the memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the multi-type material information identification and processing method as described in the first aspect of the present invention. The multi-type material information identification and processing system of the present invention has a compact structure and stable performance, and can stably execute the multi-type material information identification and processing method provided by the present invention, further improving the overall applicability and practical application capabilities of the present invention.

[0013] Beneficial effects: The present invention obtains image data of multiple types of materials through machine vision, optimizes the image data and then extracts texture features, and finally completes the recognition and processing of multiple types of material information based on DS theory combined with multiple machine model recognition results, thereby improving the grading and classification capabilities of multiple types of materials and solving the problem of fast and accurate information recognition and processing of multiple types of materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flow chart of a method for identifying and processing multi-type material information of the present invention; Figure 2 This is a framework diagram of a multi-type material information identification and processing system of the present invention; Figure 3 A schematic diagram of the structure of a multi-type material information identification and processing device provided in an embodiment. DETAILED DESCRIPTION

[0015] The specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described herein are only for illustration and are not intended to limit the present invention. In the following description, a large number of specific details are set forth in order to provide a thorough understanding of the present invention. However, it is obvious to those of ordinary skill in the art that these specific details do not need to be adopted to implement the present invention. In other examples, in order to avoid confusing the present invention, known circuits, software or methods are not specifically described.

[0016] Throughout the specification, references to "one embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment or example is included in at least one embodiment of the present invention. Therefore, the phrases "in one embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily all refer to the same embodiment or example. In addition, particular features, structures, or characteristics may be combined in one or more embodiments or examples in any suitable combination and / or subcombination. In addition, it should be understood by those of ordinary skill in the art that the figures provided herein are for illustrative purposes and that the figures are not necessarily drawn to scale.

[0017] See also Figure 1 , Figure 1 The flowchart of the method for identifying and processing information of multiple types of materials of the present invention is as follows. To solve the problem, the present invention provides a method for identifying and processing information of multiple types of materials, such as Figure 1 As shown, the method comprises the following steps: S1. Obtain a material image in Lab mode through an image acquisition device.

[0018] In the embodiment, the image acquisition device is used to acquire image data of multiple types of dry bulk materials. Therefore, the image acquisition device can be a camera fixed at a certain position near the conveyor belt, and can transmit the acquired clear image data of multiple types of dry bulk materials to the data center through the data network for subsequent operations.

[0019] Furthermore, the acquired image needs to be subjected to noise reduction processing. In the embodiment, the noise reduction processing method includes a filter-based method, a model-based method and a learning-based method. Among them, the filter-based method includes mean filtering, which replaces the original pixel value by calculating the average value of the current pixel point and the pixel points in its neighborhood, thereby achieving the effect of smoothing the image; Gaussian filtering, which applies normal distribution (Gaussian distribution) to image processing and smoothes the image; median filtering, which replaces the value of a certain point in a digital image with the median value of the values ​​of each point in an area where the point is located; and bilateral filtering, which is a compromise processing combining the spatial proximity of the image and the similarity of pixel values, while considering spatial domain information and grayscale similarity to achieve the purpose of edge-preserving denoising; and adaptive Wiener filtering, which adjusts the output of the filter according to the local variance of the image. The larger the local variance, the stronger the smoothing effect of the filter.

[0020] Model-based methods attempt to model the distribution of natural images or noise, and then use the model distribution as a priori to obtain clear images through optimization algorithms. For example, infrared weighted nuclear norm minimization based on low-rank matrix approximation, graph-based regularizers, and other methods.

[0021] Learning-based methods: In recent years, with the development of deep learning, deep network-based methods have made significant progress in the field of image denoising. This type of method learns the image prior model by training the neural network to achieve image denoising. For example, denoising convolutional neural networks, fast and flexible denoising networks, etc. This type of method has strong learning and generalization capabilities and can handle different types of noise and complex image scenes.

[0022] Furthermore, the denoised image is processed into a material image in Lab mode, which is a color model based on the international standard for color measurement established by the International Commission on Illumination (CIE). The Lab color model consists of three elements: brightness (L), and two color channels a and b. The color range of the a channel is from dark green (low brightness value) to gray (medium brightness value) to bright pink (high brightness value); the color range of the b channel is from bright blue (low brightness value) to gray (medium brightness value) to yellow (high brightness value).

[0023] S2. Construct a first image generation model, and combine the material image and the first image generation model to obtain a first image of the material image.

[0024] In the embodiment, the first image generation model constructed in step S2 satisfies the following formula: in, Represents the generated image Pixels Channel value, Represents multiple types of material images Channel average, express Channel value adjustment factor, Represents the image of multiple types of materials Pixels Channel value standard deviation, Represents the generated image Pixels Channel value, Represents multiple types of material images Channel average, express Channel value adjustment factor, Represents the image of multiple types of materials Pixels Channel value standard deviation, Represents the generated image Pixels Channel value, Represents multiple types of material images Channel average, express Channel value adjustment factor, Represents the image of multiple types of materials Pixels Standard deviation of channel values.

[0025] Specifically, each channel value of the image is calculated to regenerate a new Lab mode material image, which is the first image of the material image.

[0026] Furthermore, each channel value adjustment factor can be determined by using a group intelligence optimization algorithm such as a particle swarm algorithm, or can be determined by expert discussion and experiment. In this embodiment, .

[0027] S3. Extracting texture features of the material image from the first image.

[0028] Specifically, in step S3, extracting the texture features of the material image in the first image includes the following steps: S31, obtaining a representative pixel value of the first image; Further, the representative pixel value of the first image is obtained, which satisfies the following formula: in, Indicates The representative pixel value of a pixel point is Represents the generated image Pixels Channel value, express Channel value adjustment factor, Represents the generated image Pixels Channel value, express Channel value adjustment factor, Represents the generated image Pixels Channel value, express Channel value adjustment factor.

[0029] S32, obtaining a first coding value and a second coding value of each pixel in the first image by using a first coding algorithm, a second coding algorithm and the representative pixel value; Further, obtaining the first coding value by using the first coding algorithm comprises the following steps: Establish a nine-square grid with the pixel point as the center point; Obtaining a representative pixel value difference between the pixel point and other pixel points in the nine-square grid; The representative pixel value difference greater than the representative pixel value difference average is marked as 1, and the representative pixel value difference less than or equal to the representative pixel value difference average is marked as 0; 0 Direction statistics mark, obtaining the first encoding value; Obtaining the second encoding value by using the second encoding algorithm includes the following steps: Establishing a twenty-five-square grid with the pixel point as the center point; Get 0 in the twenty-five grids, , , , , , , The second coding value is obtained by calculating the mean of the representative pixel values ​​of the pixel points in the direction, and according to the first coding algorithm.

[0030] S33. Count the histogram vectors of the encoding values ​​of the first image to obtain texture features of the multiple types of materials.

[0031] In the embodiment, the first coding value of a certain pixel is 01001101, and the second coding value is 11100011. The histogram vector of the pixel is 0100110111100011. The histogram vectors of all the pixels of the first image are counted, which is the texture feature of the material.

[0032] S4. Based on multiple machine classification models, the texture features of multiple material image test samples are used to obtain the credibility of multiple machine classification models.

[0033] Specifically, the machine classification model described in step S4 includes a BP neural network model, a K-means classification model, a SVM classification model, a random forest model, a decision tree model and an integrated learning model.

[0034] The BP neural network model is a multi-layer feedforward network trained by the error back propagation algorithm, namely the back propagation neural network. It is also a directed graph model composed of multiple neurons, usually including an input layer, a hidden layer and an output layer.

[0035] The K-means classification model is a widely used clustering algorithm, which aims to divide n data points into k clusters so that the data points within each cluster are as similar as possible, while the data points between different clusters are as different as possible. Each cluster is represented by a centroid, which is the mean of all the points in the cluster.

[0036] The SVM classification model is an algorithm that searches for the linear classifier with the largest interval in feature space. The basic idea is to find a hyperplane to separate data of different categories so as to maximize the interval between categories, thereby achieving accurate classification of new data.

[0037] The random forest model consists of multiple decision trees, each of which is obtained by training on different samples and feature subsets. Specifically, the random forest model follows two basic principles in the construction process: data randomness and feature randomness. Data randomness refers to randomly selecting a certain number of samples from the original data set and using sampling with replacement to build a decision tree, which means that each decision tree is generated by randomly sampling the training data to ensure data diversity; feature randomness refers to randomly selecting a subset of features at each node and performing optimal division based on these features. This feature randomness helps reduce the correlation between individual decision trees and improve the generalization ability of the model.

[0038] The decision tree model makes decisions or predictions by building a tree model. The decision tree model divides data into different categories or predicts continuous value results in the form of a tree diagram. It follows a divide and conquer strategy, starting from the root node, and divides the data into smaller subsets through a series of tests until a certain termination condition such as a leaf node is reached.

[0039] The ensemble learning model is a machine learning paradigm that solves problems by combining multiple learners. The core idea is to construct multiple base learners, also known as individual learners or weak learners, and let them make independent predictions on the input data. Then, the prediction results of each base learner are combined in some way, such as voting, weighted averaging, etc. to produce a final prediction result. This method can effectively improve the prediction accuracy and robustness of the model.

[0040] Furthermore, image test samples of various types of materials are obtained, and the samples and their texture features are used as input to obtain the credibility of each machine classification model.

[0041] Furthermore, the credibility of the machine classification model satisfies the following formula: in, Indicates The credibility of the machine classification model, represents the number of machine classification models, Indicates the quantity of material type. Indicates Machine classification model recognition The success rate of the material Indicates Machine classification model recognition The success rate of the material.

[0042] S5. According to the credibility and the material image, obtain the joint credibility of the machine classification model in identifying the multiple types of materials, and complete the recognition processing of the multiple types of material information based on the joint credibility.

[0043] Specifically, in step S5, the joint credibility of the machine classification model for identifying multiple types of materials is obtained based on the credibility and the material image, satisfying the following formula: in, Indicates Machine classification model recognition The combined reliability of the materials, Indicates Machine classification model recognition The success rate of the material Indicates The credibility of the machine classification model, Indicates the uniformity of machine classification model recognition. If the machine classification model recognizes the same material image, then If the machine classification model recognizes the material image differently, , Represents the number of machine classification models.

[0044] Furthermore, the identification process of multiple types of material information is completed based on the joint credibility, including the following steps: The joint credibility is integrated by using the DS evidence theory fusion system to obtain the recognition probability of the material; An identification probability judgment rule is set, and the identification processing of multi-type material information is completed by combining the identification probability and the identification probability judgment rule.

[0045] In the embodiment, the recognition probability of the final recognized material is assumed to be ,but The identification probability judgment rule that is satisfied is one of the following: The maximum value among all material identification probabilities; is greater than a first recognition probability threshold; The differences in recognition probabilities with other materials are all greater than the second recognition probability threshold; If the above conditions are not met, the output result is "unrecognized material", and more samples are used for retraining. Furthermore, the first recognition probability threshold and the second recognition probability threshold need to be determined through multiple experiments. In this embodiment, the first recognition probability threshold is 0.82, and the second recognition probability threshold is 0.51.

[0046] See also Figure 2 In an embodiment, in order to efficiently execute the multi-type material information identification and processing method provided by the present invention, the present invention also provides a multi-type material information identification and processing system, including: an input device, an output device, a processor, and a memory, wherein the input device, the output device, the processor, and the memory are interconnected, and the memory contains program instructions, and the program instructions are used for the steps of the multi-type material information identification and processing method. The multi-type material information identification and processing system of the present invention has a compact structure and stable performance, and can stably execute the multi-type material information identification and processing method of the present invention, further improving the overall applicability and practical application ability of the present invention.

[0047] In an embodiment, the processor may be a central processing unit (CPU), which may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. An input device may be used to obtain data information. An output device may be used to output the results obtained by storing program instructions contained in a computer program in a memory provided by the present invention. The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the type of device.

[0048] In yet another alternative embodiment, see Figure 3 In order to efficiently execute the multi-type material information identification and processing method provided by the present invention, this embodiment also provides a multi-type material information identification and processing device, such as Figure 3 As shown, including: The memory 10 is used to store computer programs; the processor 20 is used to execute the computer programs to implement the above-mentioned multi-type material information identification and processing method.

[0049] Memory 10 , processor 20 , communication interface 31 and communication bus 32 . Memory 10 , processor 20 , communication interface 31 all communicate with each other via communication bus 32 .

[0050] In an embodiment, the memory 10 is used to store one or more program instructions. The memory 10 may store program instructions for implementing the following functions: acquiring a material image in Lab mode through an image acquisition device; constructing a first image generation model, combining the material image and the first image generation model to obtain a first image of the material image; extracting texture features of the material image in the first image; based on multiple machine classification models, using texture features of multi-type material image test samples, obtaining the credibility of multiple machine classification models; based on the credibility and the material image, obtaining the joint credibility of the machine classification model in identifying multiple types of materials, and completing the recognition processing of multiple types of material information based on the joint credibility.

[0051] In one possible implementation, the memory 10 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, and at least one application required for a function, etc.; the data storage area may store data created during use. In addition, the memory 10 may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include NVRAM. The memory stores an operating system and operating instructions, executable modules or data structures, or a subset thereof, or an extended set thereof, wherein the operating instructions may include various operating instructions for implementing various operations. The operating system may include various system programs for implementing various basic tasks and processing hardware-based tasks.

[0052] The processor 20 may be a central processing unit (CPU), an application specific integrated circuit, a digital signal processor, a field programmable gate array or other programmable logic device, and the processor 20 may be a microprocessor or any conventional processor, etc. The processor 20 may call a program stored in the memory 10. The communication interface 31 may be an interface of a communication module, and is used to connect with other devices or systems.

[0053] Of course, it should be noted that Figure 3 The structure shown does not constitute a limitation on the multi-type material information identification and processing device in this embodiment. In actual applications, the multi-type material information identification and processing device may include Figure 3 More or fewer components than shown, or combinations of certain components.

[0054] In the embodiment, a storage medium is also provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned multi-material information identification and processing method are implemented. The storage medium may include: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0055] In summary, the present invention obtains image data of multiple types of materials through machine vision, optimizes the image data and then extracts texture features, and finally completes the recognition and processing of multiple types of material information based on DS theory combined with multiple machine model recognition results, thereby improving the grading and classification capabilities of multiple types of materials and solving the problem of rapid and accurate information recognition and processing of multiple types of materials. Therefore, the present invention effectively overcomes the various shortcomings of the prior art and has a high industrial utilization value.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope recorded in the present invention.

Claims

1. A method for identifying and processing multi-type material information, characterized in that: The multi-type material information identification and processing method comprises the following steps: Obtain the material image in Lab mode through the image acquisition device; Constructing a first image generation model, combining the material image and the first image generation model to obtain a first image of the material image; In the first image, extracting texture features of the material image; Based on multiple machine classification models, the texture features of multiple material images are used to test samples to obtain the credibility of multiple machine classification models; According to the credibility and the material image, the joint credibility of the machine classification model in identifying multiple types of materials is obtained, and the recognition processing of the multiple types of material information is completed based on the joint credibility.

2. According to claim 1, a method for identifying and processing multi-type material information is characterized in that: The first image generation model constructed satisfies the following formula: in, Represents the generated image Pixels Channel value, Represents multiple types of material images Channel average, express Channel value adjustment factor, Represents the image of multiple types of materials Pixels Channel value standard deviation, Represents the generated image Pixels Channel value, Represents multiple types of material images Channel average, express Channel value adjustment factor, Represents the image of multiple types of materials Pixels Channel value standard deviation, Represents the generated image Pixels Channel value, Represents multiple types of material images Channel average, express Channel value adjustment factor, Represents the image of multiple types of materials Pixels Standard deviation of channel values.

3. A method for identifying and processing multi-type material information according to claim 1, characterized in that: Extracting the texture features of the material image in the first image comprises the following steps: Obtaining a representative pixel value of the first image; Obtaining a first encoding value and a second encoding value of each pixel in the first image by using a first encoding algorithm, a second encoding algorithm and the representative pixel value; The histogram vectors of the encoding values ​​of the first image are counted to obtain the texture features of the multiple types of materials.

4. A method for identifying and processing multi-type material information according to claim 3, characterized in that: The obtaining of the representative pixel value of the first image satisfies the following formula: in, Indicates The representative pixel value of a pixel point is Represents the generated image Pixels Channel value, express Channel value adjustment factor, Represents the generated image Pixels Channel value, express Channel value adjustment factor, Represents the generated image Pixels Channel value, express Channel value adjustment factor.

5. A method for identifying and processing multi-type material information according to claim 3, characterized in that: Obtaining the first coding value by using the first coding algorithm includes the following steps: Establish a nine-square grid with the pixel point as the center point; Obtaining a representative pixel value difference between the pixel point and other pixel points in the nine-square grid; The representative pixel value difference greater than the representative pixel value difference average is marked as 1, and the representative pixel value difference less than or equal to the representative pixel value difference average is marked as 0; 0 Direction statistics mark, obtaining the first encoding value; Obtaining the second encoding value by using the second encoding algorithm includes the following steps: Establishing a twenty-five-square grid with the pixel point as the center point; Get 0 in the twenty-five grids, , , , , , , The second coding value is obtained by calculating the mean of the representative pixel values ​​of the pixel points in the direction, and according to the first coding algorithm.

6. A method for identifying and processing multi-type material information according to claim 1, characterized in that: The machine classification model includes a BP neural network model, a K-means classification model, a SVM classification model, a random forest model, a decision tree model and an integrated learning model.

7. A method for identifying and processing multi-type material information according to claim 1, characterized in that: Based on multiple machine classification models, the credibility of multiple machine classification models is obtained by using the texture features of multiple material image test samples, which satisfies the following formula: in, Indicates The credibility of the machine classification model, represents the number of machine classification models, Indicates the quantity of material type. Indicates Machine classification model recognition The success rate of the material Indicates Machine classification model recognition The success rate of the material.

8. A method for identifying and processing multi-type material information according to claim 1, characterized in that: The joint credibility of the machine classification model in identifying multiple types of materials is obtained based on the credibility and the material image, satisfying the following formula: in, Indicates Machine classification model recognition The combined reliability of the materials, Indicates Machine classification model recognition The success rate of the material Indicates The credibility of the machine classification model, Indicates the uniformity of machine classification model recognition. If the machine classification model recognizes the same material image, then If the machine classification model recognizes the material image differently, , Represents the number of machine classification models.

9. A method for identifying and processing multi-type material information according to claim 1, characterized in that: The method of completing the identification process of multiple types of material information based on the joint credibility includes the following steps: The joint credibility is integrated by using the DS evidence theory fusion system to obtain the recognition probability of the material; An identification probability judgment rule is set, and the identification processing of multi-type material information is completed by combining the identification probability and the identification probability judgment rule.

10. A multi-type material information identification and processing system, characterized in that: The multi-type material information identification and processing system includes: an input device, an output device, a processor, and a memory. The input device, the output device, the processor, and the memory are interconnected. The memory includes program instructions, and the program instructions are used to execute the multi-type material information identification and processing method according to any one of claims 1 to 9.

Images