Recognition method of surface defects of multilayer solid wood composite floor with surface board being jointed board

Abstract

The invention provides an recognition method of surface defects of a multilayer solid wood composite floor with a surface board being a jointed board, which is basically the same in the classification standards, high in recognition rate and low in error rate, can realize industrial on-line detection, and can accurately recognize three types of defects of the multilayer solid composite floor: corner defects, surface fractures and splicing open joints. The recognition method comprises the following steps: performing image graying processing, image grey level transformation, image smoothing processing, image segmentation and image segmentation postprocessing on a collected RGB color image on the surface of the solid wood composite floor to extract target defects; adding characteristic values of the target defects, with defect characteristics being extracted, to a support vector machine, so as to be trained; after training, adding the characteristic values of the target defects, with defect characteristics being extracted, to the support vector machine, so as to be recognized.

Description

technical field [0001] The invention belongs to the field of surface defect recognition of solid wood composite floor, and relates to the detection of surface defect of wood floor. Background technique [0002] Due to the large number of workers in the wood flooring industry, high labor costs and the urgent need for industry transformation, it has become an inevitable trend to introduce machine vision into the wood flooring industry. On the one hand, it can reduce labor and errors caused by labor in the inspection section, and on the other hand, it can reduce manufacturing costs for enterprises. [0003] At present, the wooden floor industry is basically testing wooden floors manually. Due to the diversity and irregularity of the surface defects of solid wood composite flooring with jigsaw panels as the surface board, there are many workers in the floor testing section of the enterprise, accounting for the entire More than half of the work section, coupled with the low effi...

AI Technical Summary

Problems solved by technology

The defects of solid wood flooring study are: dead knots, live knots, bug holes and cracks, and the defects of laminate flooring study are: dry flowers, paper missing, dirt and paper cracks

However, due to the static collection of images, it cannot meet the requirements of online detection. At the same time, the stability of the algorithm is improved, and the recognition rate is relatively low.

[0010] The surface defects of laminate flooring and solid wood flooring mentioned above are all inherent defects of wood flooring. The research methods all use BP neural network to identify. The advantage is that the recognition speed is fast. The disadvantage is that compared with RBF and SOM neural network, the recognition rate is low

Images