实木地板图像差分和形态学分割算法研究

针对现有实木地板表面缺陷分割算法分割速度慢、精度低的问题，提出基于图像差分和形态学的地板表面缺陷分割方法。首先，分别对目标图像和背景图像进行求补和低通滤波运算；其次，将预处理后的2幅图像做差分运算，得到差分图像；最后，采用形态学开闭运算去除差分图像噪声并获得目标缺陷。选取活节、死节、裂纹3类缺陷样本各20个进行试验，结果表明：该研究算法对3类样本平均分割准确革分别达到92．63％、94．31％、93．88％，平均分割时间为9．583ms。

Defects Segmenting for Wood Floor Based on Image Difference and Mathematical Morphology

To solve the problem of slow speed and low accuracy of defects segmentation methods to the surface defects of wood floor, we pro- posed a novel segmenting method based on image difference and mathematical morphology. Firstly, compute the complemented image of the target image and filter the back grand image by a low pass filter. Secondly, differential operation is happened on the two pretreated images to obtain the differential image. Finally, using morphological opening and closing operation to remove the noise on the difference image and get the goal defects image. Choose sound knots, dead knots and cracks these three kinds of defect to test, and each number of the experimental subject is 20 samples. The result of the experiment reveals that： the average segmenting accuracy rate by this proposed method for each type of defect is 92.63%, 94.31% and 93.88%, and the average segmenting time is 9. 583 ms.