Recognition Research Of Micro-nano Wood Flour Mesh Based On Microscopic Image Processing

As a kind of new environmental material, micro-nano wood flour whose application is decided by mesh has been applied more and more broadly. Traditional mesh recognition method cannot meet the recognition demand of current produce that is of high-efficient and fast speed because of low work efficient and too much operation among wood flour producing. Therefore, it is a fast and effective mesh recognition method that is the key to realizing automatic monitoring mesh in the wood flour producing industry.The subject of this paper is supported by "Harbin Research Special Funds of Science and Technology Innovative Talents" and is part of "wood ultrafine crushing air classifier modeling and computer mesh recognition". Its task is micro-nano mesh recognition based on microscopic image, solves the problems during image processing and recognition, and proposes proper algorithms so as to realize mesh detection.During image processing, in view of the low contrast and high noise shooting environment, this paper firstly transforms color space from RGB to HSV, and carries out the binarization based on S component; then in order to remove noise and impurity, improve the image by hole filling, math morphology technology and so on; finally detect the edge of the target by the modified morphological multiscale operator to gain wood floor’s contour successfully.During wood flour mesh recognition, according to the further research on particle size test approach in the field of particle size measurement, this paper proposes to use rectangle fitting algorithm, gains mesh by unit conversion based on morphological characteristics and analyzes particle size to facilitate the processing of the data. In accordance with the actual demand for mesh recognition, a micro-nano wood flour mesh recognition system based on modular approach has been designed and developed as well as designs mesh recognition program based on wood flour features.The experimental results has demonstrated that the proposed algorithm do better in wood flour microscopic image processing and mesh recognition. It has higher reliability and appropriateness, and has great significance for promoting wood flour mesh intelligent detection.