文档介绍：Anhui Agri Sci Bull 2022 28 07 97
安徽农学通报， ， ， ， inhai
（
Agricultural Technology Development Co., Ltd, Yuxi 650224, China; 3Key Laboratory of Advanced Perception and
Intelligent Control of High-end Equipment of Ministry of Education, Anhui Polytechnic University, Wuhu 241000,
China; 4School of Electrical Engineering, Anhui Polytechnic University, Wuhu 241000, China
）
In view of the discrete, disorderly distribution and complicated edge details of rose leaf powdery mildew
Abstract:
disease spots, traditional image segmentation algorithms are difficult to extract the disease spots, and a new leaf dis⁃
ease spot extraction algorithm based on wavelet filtering and adaptive mean threshold segmentation WT-AAT is
（ ）
proposed. First, in order to remove the original image noise, the image is decomposed by two-layer wavelet using
two-dimensional discrete Haar wavelet transform, and the noise coefficient is automatically extracted according to
the high and low frequency sub-images, and then the filter threshold is set; then, zero the noise figure above the
threshold, and