CN-122016580-A - High-raw water electro-gravel aggregate particle size online detection system based on machine vision

Abstract

The invention discloses a machine vision-based high-raw water electro-gravel aggregate particle size online detection system which comprises an image acquisition module, a protection device, a supporting structure, an image processing module, a data analysis module and a control module, wherein the image acquisition module comprises an industrial camera and a light source, the protection device and the supporting structure are arranged above a belt conveyor, the image processing module is used for image preprocessing, the example segmentation based on Mask R-CNN and the particle size calculation of an equivalent elliptic Feret short diameter, the data analysis module is used for calculating particle size distribution and super/super diameter rate in real time, and the control module is used for realizing closed-loop control in linkage with crushing and screening equipment.

Inventors

• ZHOU HUA
• ZHANG XI
• DENG YONGJUN
• ZHAO GUOQIANG
• ZHU XINJIE
• DAI JIHU
• LV YU
• MAO JIA
• ZHOU GANG
• HUANG QUN
• LIU YIFEI
• PANG BOHUI
• ZHANG YANG
• WANG DAOMING
• Li haida
• MA HAIZHONG

Assignees

• 华能澜沧江上游水电有限公司
• 中国水利水电第七工程局有限公司
• 华能澜沧江水电股份有限公司
• 中国科学院上海高等研究院

Dates

Publication Date

20260512

Application Date

20251224

Claims (10)

1. 1. High raw water electric sand aggregate particle size on-line measuring system based on machine vision, characterized by including: the image acquisition module comprises an industrial camera and an LED light source, is fixedly arranged above the belt conveyor and is used for acquiring images of sand aggregates in real time in a belt running state; the protection and installation module comprises a protection device and a supporting structure, wherein the protection device adopts low-temperature resistant, dustproof, waterproof and ultraviolet-proof designs, the image acquisition module is packaged, and the supporting structure adopts an anti-vibration aluminum alloy frame and is used for fixing the image acquisition module and the protection device; The image processing and analyzing module comprises an image processing module and a data analyzing module, is connected with the image acquisition module, and is used for preprocessing an acquired image, dividing a target and calculating the particle size, and the data analyzing module is used for calculating the particle size distribution, the diameter exceeding rate, the diameter losing rate and the diameter intermediate rate of the aggregate of the sand in real time and automatically identifying the mixing and misplacing conditions; The feedback and control module is connected with the image processing and analyzing module and is used for being linked with the crushing and screening equipment, and the production parameters are adjusted based on the detection result to form closed-loop control; the data storage and display module is used for presenting in the forms of a report form, a screen residual curve and a real-time curve and storing the report form, the screen residual curve and the real-time curve into a database; the edge calculation server is arranged in the control room, is in communication connection with the image processing module and the data analysis module, and is used for executing an image processing algorithm and data calculation and avoiding exposure to the plateau outdoor environment.
2. 2. The machine vision-based online detection system for the particle size of high-raw water electro-gravel aggregate is characterized in that an industrial camera in the image acquisition module is a high-speed industrial camera, and clear images can be captured under a belt high-speed running state.
3. 3. The machine vision-based online detection system for the particle size of the high-raw water electro-gravel aggregate is characterized in that an LED light source in the image acquisition module is a uniform illumination light source and is used for providing compensation illumination under the condition of high-altitude strong light, so that image shadows and light reflection are reduced.
4. 4. The machine vision-based online detection system for the particle size of the high-raw water electro-gravel aggregate is characterized in that the protection device comprises a sealed shell, a temperature compensation unit and an ultraviolet filter cover and is used for ensuring that internal equipment normally operates in a high-altitude low-temperature, strong ultraviolet and dust environment.
5. 5. The machine vision-based high-raw water electro-gravel aggregate particle size online detection system according to claim 1, wherein the image processing module comprises: The preprocessing unit is used for denoising, removing motion blur, balancing brightness and correcting gamma for the image so as to adapt to high altitude strong light and dust environment; The target segmentation unit is used for identifying and segmenting the sand particles one by one based on a Mask R-CNN example segmentation model, so that particle overlapping or adhesion is avoided; and the particle size calculation unit is used for carrying out morphological fitting on the segmented particles, calculating the equivalent elliptic Feret short diameter and converting the equivalent elliptic Feret short diameter into the actual physical particle size through the calibration plate.
6. 6. The machine vision-based online detection system for the particle size of the high-raw water electro-gravel aggregate is characterized in that the system collects a gravel aggregate image in real time under a plateau environment through an industrial camera and an LED light source which are arranged above a belt conveyor, then denoising, motion blur removal, brightness equalization and gamma correction are carried out on the collected image to eliminate the influence of high-altitude strong light and dust, then an example segmentation model based on Mask R-CNN is adopted to identify and segment the gravel particles one by one, morphological fitting is carried out on the segmented particles, equivalent elliptic Feret short diameters are calculated, the scaling is carried out to actual physical particle sizes through a calibration plate, then particle size distribution, super diameter rate, inferior diameter rate and median diameter rate are calculated in real time, whether mixing or misplacement exists is judged, finally a histogram, a screen surplus curve and a real-time data report are generated, and the histogram is stored in a database, and when the super diameter rate or inferior diameter rate exceeds a set threshold value, an alarm is triggered, and parameters of a crusher and a sieving machine are automatically adjusted, and closed-loop control is achieved.
7. 7. The machine vision-based online detection system for the particle size of the high-raw water electro-gravel aggregate is characterized by further comprising contrast enhancement operation on the acquired image, wherein an adaptive histogram equalization method is adopted for improving the definition of the image under the conditions of high altitude and low illumination.
8. 8. The machine vision-based online detection system for the aggregate particle size of the high-altitude electro-gravel is characterized in that the Mask R-CNN model is trained by a gravel image data set collected in a plateau environment, and gravel particles under dust, shadow and particle adhesion conditions can be identified.
9. 9. The machine vision-based online detection system for the particle size of the high-raw water electro-gravel aggregate according to claim 6, wherein the particle size calculation specifically comprises the steps of fitting an equivalent ellipse to each segmented gravel particle, calculating the Feret short diameter of the ellipse as the representative particle size of the particle, and finally converting the pixel size into the actual physical size through a pre-calibrated pixel-physical size proportional relationship.
10. 10. The machine vision-based high-raw water electro-gravel aggregate particle size online detection system is characterized in that when the super diameter rate exceeds a threshold value, a crusher discharge hole is automatically adjusted, and when the super diameter rate exceeds the threshold value, the vibration frequency or the inclination angle of a screening machine is automatically adjusted so as to optimize aggregate grading in real time.

Description

High-raw water electro-gravel aggregate particle size online detection system based on machine vision Technical Field The invention relates to the technical field of sand aggregate production detection and quality control, in particular to a high-raw water electro-sand aggregate particle size online detection system based on machine vision. Background When a large hydropower station is built in a plateau area, a sand and stone processing system is often required to be built nearby a dam site so as to meet the continuous supply requirement of mass concrete on aggregate. The grain size distribution of the sand aggregate is directly related to the mixing ratio, compactness, temperature control effect and structural crack resistance of the concrete. Under the special circumstances of the highland, the traditional mode has four defects, namely, the air in the highland is thin, the ultraviolet rays are strong, the day-night temperature difference is large, frequent dust raising and extreme weather are involved, the manual site sampling difficulty is high, the safety is low, the sample representativeness is easy to be disturbed by the environment, the laboratory screening process is complex, the consumed time is long, the real-time fluctuation of the aggregate particle size in the production line is difficult to be reflected in time, the concrete grading deviation is easy to be caused, the third, the construction of the hydropower station dam needs long-period and large-scale continuous feeding, the laboratory detection cannot meet the high-frequency detection requirement, and the fourth, the existing detection means lacks an effective data communication and intelligent linkage mechanism with production equipment such as crushing and screening, and the real-time closed-loop regulation of production parameters is difficult to be realized, and the long-term stability of the aggregate grading is difficult to be influenced. Therefore, in order to solve the problems, the invention provides a machine vision-based online detection system for the particle size of the high-raw water electro-gravel aggregate Disclosure of Invention In order to solve the problems of high safety risk, poor real-time performance, insufficient scale requirement and insufficient linkage of a production system in manual detection, the invention provides a high-raw water electro-sandstone aggregate particle size on-line detection system based on machine vision, and the real-time and accurate detection of the sandstone aggregate particle size is realized through image acquisition and intelligent analysis means, and closed-loop control is formed with the production line, so that the continuous stability of the aggregate quality is ensured. The technical scheme of the invention is that the high-raw water electro-gravel aggregate particle size online detection system based on machine vision comprises the following components: The image acquisition module comprises an industrial camera and an LED light source, is fixedly arranged above a belt conveyor and is used for acquiring images of sand and gravel aggregate in real time under the running state of the belt, wherein the frame rate of the industrial camera is not lower than 30fps, and is provided with a polarizer to inhibit the reflection of the aggregate surface; The protection device adopts the design of low temperature resistance, dust prevention, water prevention and ultraviolet prevention, and encapsulates the image acquisition module; The support structure adopts an anti-vibration aluminum alloy frame and is used for fixing the image acquisition module and the protection device; The image processing module is connected with the image acquisition module and is used for preprocessing the acquired image, dividing the target and calculating the particle size; the data analysis module is connected with the image processing module and used for calculating the particle size distribution, the super diameter rate, the diameter-less rate and the intermediate diameter rate of the sand aggregates in real time and automatically identifying the mixing and error conditions; The feedback and control module is connected with the data analysis module and is used for being linked with the crushing and screening equipment, and the production parameters are adjusted based on the detection result to form closed-loop control; the data storage and display module is used for presenting in the forms of a report form, a screen residual curve and a real-time curve and storing the report form, the screen residual curve and the real-time curve into a database; the edge calculation server is arranged in the control room, is in communication connection with the image processing module and the data analysis module, and is used for executing an image processing algorithm and data calculation and avoiding exposure to the plateau outdoor environment. Preferably, the industrial camera in the image acquisition module is a high-spee