Study on Plant Pest Images Identification Based on Lifting Wavelet Transform

摘要

Aiming at the weak noise signals the existence of plant diseases and insect pest images, pathological activity regulation of identification of the plant, so that between the signal molecules exist in plants can regulate mutually, collaborative work. Therefore, identification of weak signal molecules in plants is significant to the study of plant life activities. Taking corn pest images as the research object, using the identification method of lifting wavelet transform, combined with image identification technology, calculated the original plant diseases and insect pest images by not detect the break point of signal. Simulation results show that, the analysis of lifting wavelet of plant disease image identification technology reliability is about 71.65%; the accuracy of edge detection is about 76.21%. The operation speed of this algorithm is fast, easy for hardware implementation, provides an effective method for plant disease images identification.