基于机器视觉和小波分析的农田害虫识别系统-控制理论与控制工程专业论文.docxVIP

郑卅l大学工学硕士论文颜色特征压缩为一维，因此共形成了七个有效特征。 郑卅l大学工学硕士论文 颜色特征压缩为一维，因此共形成了七个有效特征。 6．农田害虫的识别与分类。结合本文研究对象，针对提取的七个有效特征， 设计并训练了BP网络分类器；针对传统BP算法易陷入局部极小和收敛速度慢的 问题，分析了各种改进方法。 用Visual c++6．0语言实现了上述各环节的功能，开发了识别系统软件包，与 研制的硬件装置相配合，分别对茶斑蛾、褐边绿刺蛾、栗六点天蛾、甜菜夜蛾、 玉米螟、棉铃虫、粘虫等8种常见的农田害虫进行测试，识别率达到了85．7％。 在农田害虫自动识别方面，还没有类似的报导，本论文初次试验，在软硬件设 计上还存在很多缺陷，能达到85．7％的识别率基本令人满意。由于我国省、市、县 各级都设有植保站，农田害虫检测预报是重要工作，因此本系统有很大的市场空 间。今后应在识别速度方面、特征提取方面、硬件设计方面有待改进，提高害虫 识别率，尽快把该系统推向市场。 关键词 小波分析，图像处理，特征提取，模式识别，机器视觉，农田害虫 蕊孚糖器筏NN4,渡势N瓣寝嚣器般涎蕊系缆s蝌。 蕊孚糖器筏NN4,渡势N瓣寝嚣器般涎蕊系缆 s蝌。 目s群。_鼎 !删1螂 鲁 鼎帮#=o!∞鼎螂爨景蜘群燃嘲耀墨糍燃糟鼎鼍燃趟删糕喜舅咧螋掌喜黑鼎嗍燃篇置!_e獬 A嚣STRAC善 China i§a main agricultural country,Insect pest detection 8糕d prediction in agriculture are a raain task of alI levels pest confroI services、A￡the present time the prediction method of attracting pest with black light and recognizing mid cotmting by mall is generally adopted．There sre some serious shortages such as bad recognition accuracy and low efficiency,It reduces seriously accuracy and timeliness of prediction and is disadvantage濂guiding insect disease prevention．Therefore,this paper researched image recognition technology based 011 computer vision and wavel褙 analysis This technology will atitomatically detect and predict insect pests． We designed and made the attraction device to attract agricultural pests，obtained agricultural pests’images with the color camera,processed images罄8s《on wavelet analysis On the basis of these,We emphasized濑extracting effective features,put forward recognizing pests’classes with pests’colors and texture features，and succeeded in extracting酝e efficient features such as color features,wave image edge moments features and SO On。Then We selected features,inpu犍ed into neural network classifier, recognized pattern,presented detection results。 Based on mentioned above jprinciples，the paper finished the following contents against agficulturN pests． 1．To design hardware device．We attracted insect pests with corrlmon 20W black light which was used by pest co