The Detection of Impurity Content in Machine-Picked Seed Cotton Based on Image Processing and Improved YOLO V4

The detection of cotton impurity rates can reflect the cleaning effect of cotton impurity removal equipment, which plays a vital role in improving cotton quality and economic benefits. Therefore, several studies are being carried out to improve detection accuracy. Image processing technology is increasingly used in cotton impurity detection, in which deep learning technology based on convolution neural networks has shown excellent results in image classification, segmentation, target detection, etc. However, most of these applications focus on detecting foreign fibers in lint, which is of little significance to the parameter adjustment of cotton impurity removal equipment. For this reason, our goal was to develop an impurity detection system for seed cotton. In image segmentation, we propose a multi-channel fusion segmentation algorithm to segment the machine-picked seed cotton image. We collected 1017 images of machine-picked seed cotton as a dataset to train the detection model and tested and recognized 100 groups of samples, with an average recognition rate of 94.1%. Finally, the image segmented by the multi-channel fusion algorithm is input into the improved YOLOv4 network model for classification and recognition, and the established V–W model calculates the content of all kinds of impurities. The experimental results show that the impurity content in machine-picked cotton can be obtained effectively, and the detection accuracy of the impurity rate can increase by 5.6%.

Using FCAW with Electrodes Based on Fe-Ti-Mo-B-C System for Increasing of Durability of Junk Removal Tools

In this work were analyzed factors and working conditions that leads to the wearing of junk mills tools that are a part of junk removal equipment used in drilling and workover of borehole. Such equipment is a part of oil and gas industry and work under condition of intense abrasive wearing with increased pressures and cyclic loads. Was established that traditional hardfacing materials based on the Fe-Cr-C system are not effective for improvement of abrasion resistance of elements of such equipment due to their low crack resistance and low hardness of chromium carbides. The aim of this work was to increase a durability of that equipment by using of flux cored electrodes with reaction components of pure metal powders, which leads to forming the fine-grained structure with increased hardness. Powders of Ti, Mo, B4C and their combinations were used. Structures of the hardfacing coatings were investigated by method of metallography, scanning electron microscopy (SEM). Abrasion wear tests were held under condition of fixed and non-fixed abrasion. Using of pure metal powders led to formation of a fine-grained structure with grains of Mo2FeB2 that forms around TiC, which work as modifier. It was investigated that the researched material based on Fe-Ti-Mo-C-B system that was used for increasing the wear resistance of junk mills led to increasing of the TBO period in 1.5-1.6 times comparing with serial hardfacing materials based on tungsten.

Numerical Simulation of Microparticles Motion in Two-Phase Bubbly Flow

The radioactive aerosol removal equipment is used as one of the safety systems of nuclear reactors. In this equipment, microparticles of aerosol are removed through gas-liquid interfaces of two-phase flow. The mechanism related to the removal of microparticles through the gas-liquid interface is not precise; a numerical evaluation method of performance of aerosol removal equipment is not realized. Then, we have started to construct a numerical simulation method to simulate the removal of microparticles through gas-liquid interfaces. In this simulation method, a detailed two-phase flow simulation code TPFIT is used as the basis of this method. TPFIT adopts an advanced interface tracking method and can simulate interface movement and deformation directly. Also, to simulate the movement of particles, the Lagrangian particle tracking method is incorporated. By combining the interface tracking method, and the Lagrangian particle tracking method, the interaction between interfaces and microparticles can be simulated in detail. To solve the Lagrangian equations of particles, fluid properties and fluid velocity surrounding aerosol particles are evaluated by considering the relative position of particles and gas-liquid interface, to simulate particle movement near the interface. In this paper, we show an outline and preliminary results of this simulation method.

Historical Relationship between Snow Depth and Damaged Area in South Korea

<p>Recently, snow disasters have been increased in South Korea due to the unexpected heavy snow in a region where winter gives little snow. For instance, 10 people were dead by the collapsed roof due to the unusual heavy snow. Many local governments do not have enough snow removal equipment because of little snow in winter season. Therefore, it has been important to estimate the amount of snow damage to prepare heavy snow disaster. There are not many researches to estimate damage of snow disaster in South Korea. In this study, historical snow damage data from 1994~2018 recorded in National Disaster Report were used to predict the future snow disaster damage using a statistical equation. However, it was not easy to predict the amount of snow damage when the heavy snow is happened in the area where no snow during the winter in history. Therefore, the relationship between the snow depth and damaged area were analyzed using the historical damage data. Principal multiple regression method was applied to develop the snow damage estimation function using the damaged area. The developed model could be applied to plan the budget for the snow removal equipment or snow damage reduction.</p><p> </p><p><strong>Acknowledgement</strong>:</p><p>This work was supported by Korea Environment Industry & Technology Institute (KEITI) through Intelligent Management Program for Urban Water Resources Project, funded by Korea Ministry of Environment(MOE) (2019002950002).</p><p> </p>