Phosphorite grade sorting method and equipment based on FPGA and ARM architecture
By using the hyperspectral data acquisition and regression model calculation method with FPGA and ARM architecture in phosphate grade sorting, the problem of low accuracy and speed of phosphate grade sorting in the existing technology is solved, and efficient and accurate phosphate grade sorting is achieved.
Patent Information
- Application Number
- CN202510165066.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-14
- Publication Date
- 2025-05-23
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing phosphate ore grade sorting methods have complex processes, high cost, slow speed and low accuracy, which cannot meet the market's diversified demand for phosphate ore grade.
Using a method based on FPGA and ARM architecture, the hyperspectral data of ore is collected in real time through a hyperspectral camera, and a regression model is constructed for grade calculation and sorting, realizing quantitative measurement and accurate sorting.
It improves the accuracy and speed of phosphate grade sorting, can meet the diversified demands for phosphate grades in different markets, and reduces costs and complexity.
Smart Images
Figure CN120028361A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of phosphate ore sorting, and in particular relates to a phosphate ore grade sorting method and equipment based on FPGA and ARM architecture. Background Art
[0002] The grade of phosphate ore generally refers to the content of phosphorus pentoxide in phosphate ore. However, the grades of phosphate ore are often different during mining, and the market requirements for the grade of phosphate ore vary due to many factors. Therefore, the grade of phosphate ore needs to be continuously measured and sorted according to different grades to meet different market application requirements.
[0003] The existing domestic phosphate rock grade measurement mainly relies on chemical methods. The phosphate rock needs to be sent to a ball mill to be ground into slurry, and then dried and subjected to a series of replacement reactions through chemical agents, and finally the phosphorus pentoxide content is detected.
[0004] However, this method has a complex process flow, high cost, and slow speed, and is generally only used for small-scale sampling. Traditional X-ray and visible light sorting of phosphate ore can only make rough qualitative distinctions, cannot perform quantitative calculations, and has a low accuracy rate. Summary of the invention
[0005] The purpose of the present invention is to provide a phosphate ore grade sorting method based on FPGA and ARM architecture, which collects hyperspectral data of ore in real time through a hyperspectral camera, and performs grade calculation and sorting through a constructed regression model, thereby solving the problems of slow speed and low accuracy in existing phosphate ore grade sorting.
[0006] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:
[0007] The present invention is a phosphate ore grade sorting method based on FPGA and ARM architecture, comprising the following steps:
[0008] Step 1: Build a regression model; collect phosphorus grade data at various locations on the surface of ore with known grade using a handheld X-ray fluorescence analyzer as sample data to implement model training;
[0009] Step 2: Use a hyperspectral camera to collect hyperspectral data of the ore on the transmission line in real time; the spectral range is 450nm to 2700nm, and is divided into several narrow bands, and the signal of each narrow band is converted into band reflectivity data;
[0010] Step 3: Convert the hyperspectral data into pseudo-color data, accurately locate all ores through the detection network, and then only calculate the located ores;
[0011] Step 4: Establish the reflectance data of several bands into multi-channel image data, and perform grade regression on each pixel through the regression model, so as to calculate the phosphorus distribution on the ore surface and realize grade sorting.
[0012] As a preferred technical solution of the present invention, at least 100 images are required for model training in step 1, and each image is annotated with at least 500 points.
[0013] As a preferred technical solution of the present invention, the step 2 also includes preprocessing the collected hyperspectral data, including but not limited to removing noise, correcting radiation and enhancing contrast.
[0014] As a preferred technical solution of the present invention, the preprocessing method includes a mean filtering, a median filtering or a Gaussian filtering denoising method, as well as a radiation correction and normalization method.
[0015] As a preferred technical solution of the present invention, the number of narrow bands in step 2 is 384.
[0016] As a preferred technical solution of the present invention, the step three also includes extracting features from the hyperspectral data, and the extracted features include but are not limited to spectral features, texture features and shape features.
[0017] As a preferred technical solution of the present invention, the spectral features are extracted by calculating one or more of the mean, standard deviation, slope, and peak parameters of the spectral curve.
[0018] As a preferred technical solution of the present invention, the texture features are extracted by calculating the gray level co-occurrence matrix or the local binary pattern method.
[0019] As a preferred technical solution of the present invention, the shape feature is extracted by edge detection or morphological analysis method.
[0020] A phosphate ore grade sorting device based on FPGA and ARM architecture, comprising a feeding system, a hyperspectral system and a sorting execution system, characterized in that it also comprises an FPGA processing unit and an ARM processing unit, the FPGA processing unit and the ARM processing unit communicate with each other through a data bus to realize data transmission and processing; wherein the sorting device applies the above-mentioned phosphate ore grade sorting method based on FPGA and ARM architecture.
[0021] The present invention has the following beneficial effects:
[0022] The present invention utilizes the different spectral characteristics of different elements, collects hyperspectral data of ore on the transmission line in real time through a hyperspectral camera, and uses a regression model to perform grade regression on each pixel point, thereby calculating the phosphorus distribution on the ore surface, achieving quantitative measurement of the phosphate ore grade, and providing a basis for subsequent sorting, thereby effectively improving the overall sorting accuracy and sorting speed.
[0023] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.
[0025] Figure 1 This is a flow chart of a phosphate ore grade separation method based on FPGA and ARM architecture of the present invention;
[0026] Figure 2 It is a structural schematic diagram of a phosphate ore grade sorting equipment based on FPGA and ARM architecture;
[0027] In the accompanying drawings, the components represented by the reference numerals are listed as follows:
[0028] 1-Feeding system, 2-High light system, 3-Sorting execution system, 4-Transmission system, 5-Lighting system, 6-Material separation system. DETAILED DESCRIPTION
[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0030] In the description of the present invention, it should be understood that the terms "opening", "upper", "lower", "thickness", "top", "middle", "length", "inside", "all around" and the like indicating orientation or positional relationship are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the components or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.
[0031] Embodiment 1
[0032] See also Figure 1 As shown, the present invention is a phosphate ore grade sorting method based on FPGA and ARM architecture, comprising the following steps:
[0033] Step 1: Build a regression model; collect phosphorus grade data at various locations on the surface of ore of known grade as sample data through a handheld X-ray fluorescence analyzer to implement model training; and at least 100 images are required for model training, and each image is annotated with at least 500 points. At the same time, deep learning neural networks, support vector machines, random forests, etc. are used to classify the extracted features to determine the type and quality of the phosphate ore.
[0034] Deep learning neural networks have powerful feature learning and classification capabilities, and can automatically learn features from hyperspectral data and perform accurate classification. Algorithms such as support vector machines and random forests also have high classification accuracy and stability. By training a large number of phosphate rock samples, the parameters of the classification model are optimized to improve the accuracy and generalization of classification.
[0035] Step 2: With a full-band light source, a hyperspectral camera is used to collect hyperspectral data of the ore on the transmission line in real time. The spectral range is 450nm to 2700nm and is divided into 384 narrow bands. The signal of each narrow band is converted into band reflectivity data.
[0036] In addition, the collected hyperspectral data are preprocessed, including but not limited to noise removal, radiation correction and contrast enhancement. The preprocessing methods include mean filtering, median filtering or Gaussian filtering denoising methods, as well as radiation correction and normalization methods. Preprocessing can improve the quality and reliability of the data and provide a better basis for subsequent feature extraction and classification.
[0037] Step 3: Convert the hyperspectral data into pseudo-color data, accurately locate all the ores through the detection network, and then only calculate the located ores. At the same time, it also includes feature extraction from the hyperspectral data, and the extracted features include but are not limited to spectral features, texture features, and shape features.
[0038] Among them, spectral features are extracted by calculating one or more of the mean, standard deviation, slope, and peak parameters of the spectral curve; texture features are extracted by calculating grayscale co-occurrence matrix or local binary pattern; shape features are extracted by edge detection or morphological analysis. Feature extraction can reduce data dimensions and improve subsequent sorting efficiency and accuracy.
[0039] Step 4: The 384-band reflectance data are established into multi-channel image data, and the grade of each pixel is regressed through the regression model to calculate the phosphorus distribution on the ore surface and achieve grade sorting.
[0040] Among them, the model input is 384-dimensional emissivity data of each point, and the output is the grade data of each point. By locating the hyperspectral information of the ore, 384 band data are integrated to establish 384-channel image 3D data, and a 384-channel convolutional regression network is constructed to train and learn multi-channel image data to calculate the phosphorus grade value of each part of the ore surface.
[0041] Therefore, a 384-channel image is established using the full-band spectral reflectance data, and a 384-channel convolution regression model is established by making full use of the characteristics of the reflectance of different elements in different bands, so as to achieve quantitative calculation of the phosphorus content in various parts of the ore surface, effectively improving the overall sorting accuracy and speed.
[0042] Embodiment 2
[0043] A phosphate ore grade sorting device based on FPGA and ARM architecture comprises a feeding system 1, a hyperspectral system 2, a sorting execution system 3, a transmission system 4, a lighting system 5 and a material sorting system 6, and also comprises an FPGA processing unit and an ARM processing unit. The FPGA processing unit and the ARM processing unit communicate with each other through a high-speed data bus to realize data transmission and processing.
[0044] The hardware control of the sorting equipment is mainly implemented by FPGA and ARM, which are responsible for controlling and coordinating the various hardware modules of the equipment.
[0045] For example, the sorting execution system 3 includes a high-frequency valve and an air jet device. When the sorting result determines that a certain phosphate ore particle does not meet the requirements, the FPGA+ARM architecture quickly executes the valve blowing instruction, drives the high-frequency valve to operate, and uses the air jet device to blow the particle away from the main material flow.
[0046] The high-frequency valve has the characteristics of fast response and high precision, which can realize the accurate sorting of phosphate ore particles. The jet device can adjust the jet pressure and direction according to different sorting requirements to ensure the sorting effect.
[0047] Software control is mainly implemented by the operating system and application programs running on ARM, which are responsible for managing and controlling the equipment's parameter settings, data acquisition, processing, classification, and sorting operations. By utilizing the high-speed processing capabilities of the FPGA+ARM architecture, it can realize the rapid acquisition, preprocessing, and classification of hyperspectral data, thereby improving sorting efficiency.
[0048] The sorting equipment applies the phosphate ore grade sorting method based on FPGA and ARM architecture in Example 1. For example, phosphate ore samples of known grade of different types and qualities are prepared and inputted by the feeding system 1 of the sorting equipment to train the classification regression model, and the parameters of the equipment are adjusted according to the actual situation according to the operating status and sorting effect of the monitoring equipment to optimize the sorting effect.
[0049] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.
[0050] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.
Claims
1. A phosphate ore grade sorting method based on FPGA and ARM architecture, characterized in that: The following steps are involved: Step 1: Build a regression model; The phosphorus grade data of each position on the surface of the ore with known grade is collected by handheld X-ray fluorescence analyzer as sample data to realize model training; Step 2: Use a hyperspectral camera to collect hyperspectral data of the ore on the transmission line in real time; the spectral range is 450nm to 2700nm, and is divided into several narrow bands, and the signal of each narrow band is converted into band reflectivity data; Step 3: Convert the hyperspectral data into pseudo-color data, accurately locate all ores through the detection network, and then only calculate the located ores; Step 4: Establish the reflectance data of several bands into multi-channel image data, and perform grade regression on each pixel through the regression model, so as to calculate the phosphorus distribution on the ore surface and realize grade sorting.
2. The phosphate ore grade separation method based on FPGA and ARM architecture according to claim 1 is characterized in that: In step 1, at least 100 images are required for model training, and each image is annotated with at least 500 points.
3. The phosphate ore grade separation method based on FPGA and ARM architecture according to claim 1 is characterized in that: The step 2 also includes preprocessing the collected hyperspectral data, including but not limited to removing noise, correcting radiation and enhancing contrast.
4. The method for phosphate ore grade separation based on FPGA and ARM architecture according to claim 3 is characterized in that: The preprocessing method includes a mean filter, a median filter or a Gaussian filter denoising method, and a radiation correction and normalization method.
5. The method for phosphate ore grade separation based on FPGA and ARM architecture according to claim 1, characterized in that: The number of narrow bands in step 2 is 384.
6. The method for phosphate ore grade separation based on FPGA and ARM architecture according to claim 1, characterized in that: The step three also includes extracting features from the hyperspectral data, and the extracted features include but are not limited to spectral features, texture features and shape features.
7. The method for phosphate ore grade separation based on FPGA and ARM architecture according to claim 6, characterized in that: The spectral feature is extracted by calculating one or more of the mean, standard deviation, slope, and peak parameters of the spectral curve.
8. The method for phosphate ore grade separation based on FPGA and ARM architecture according to claim 6, characterized in that: The texture features are extracted by calculating the gray level co-occurrence matrix or the local binary pattern method.
9. The method for phosphate ore grade separation based on FPGA and ARM architecture according to claim 6, characterized in that: The shape features are extracted by edge detection or morphological analysis methods.
10. A phosphate ore grade sorting device based on FPGA and ARM architecture, comprising a feeding system, a hyperspectral system and a sorting execution system, characterized in that: It also includes an FPGA processing unit and an ARM processing unit, wherein the FPGA processing unit and the ARM processing unit communicate with each other via a data bus to achieve data transmission and processing; Wherein, the sorting equipment applies the phosphate ore grade sorting method based on FPGA and ARM architecture as described in any one of claims 1 to 9.