Image-based gravel particle size detection and analysis system

Abstract

The invention discloses an image-based gravel particle size detection and analysis system, which comprises a rack, discharging equipment, image acquisition equipment, light supplementing equipment, image processing equipment, a display, a sliding block and a recovery box, segmenting the sandstone image by the image processing equipment comprises the following sub-steps: firstly, taking ResNet based on a feature pyramid network (FPN) as a feature extraction network, secondly, proposing a self-adaptive ROIAlign module, mapping each suggestion box output by the RPN to all feature levels of the FPN. A multi-scale feature is adopted to obtain a maximum value through a parameter layer for fusion, finally, an IoU prediction branch is added to improve the positioning performance of a network, moreaccurate segmentation is achieved, an NMS algorithm based on positioning confidence is provided to preferentially reserve a detection frame with high positioning precision to be provided for a targetdetection branch and a mask branch. Segmentation precision is further improved. The method can be applied to real-time on-line analysis of the grain size of the sandstone in a factory.

Description

[technical field] [0001] The invention relates to an image-based particle size detection and analysis system for sand and gravel particles. [Background technique] [0002] The coarse aggregate grading ratio of concrete is especially important for the working performance of the concrete mixture and the strength and hardness shrinkage of the hardened concrete. The particle size distribution detection of concrete aggregate is indispensable in the transportation industry and construction industry. In the link, the aggregate must be used in production practice after ensuring that the particle size distribution of the concrete is up to standard. However, the traditional mechanical sieving method uses mechanical shaking to make concrete aggregates pass through sieve holes of different apertures. This method is noisy, high in cost, slow in detection speed, and difficult to achieve high accuracy. [0003] The present invention is produced based on the above problems. [Content of t...

AI Technical Summary

Problems solved by technology

However, the traditional mechanical sieving method uses mechanical shaking to make the concrete aggregate pass through the sieve holes of different apertures. This method has high noise, high cost, slow detection speed, and it is difficult to achieve high accuracy.

Images