摘要
为了智能化农业的发展,提高农作物病害诊断水平,及时采取防治措施,提出一种基于图像处理技术与Android手机的病害智能诊断系统的设计方法.系统不受时间和地域限制,用户可以在具备网络覆盖的地方将采集到的病害图片发送至服务器端.服务器端接收到病害图片后,在HSI(horizintal situtation indicator)颜色空间对病害图像分割,利用颜色矩和灰度共生矩阵来提取病害颜色和纹理特征参数,并将优选的特征参数输入支持向量机识别,识别结果反馈给客户端.实验结果表明:该系统可以准确高效地识别出病害种类,有较强的实用性和推广应用前景.
To develop the intelligent agriculture,improve the diagnostic level of crop diseases and take timely prevention and control measures,a design method of a disease intelligent diagnostic system based on image processing technology and Android phone was proposed.As the system was not restricted by time and region,users could send the collected disease images to the server anywhere where there was network coverage.After the server received the disease images,the image segmentation was carried out in HSI color space,color moment and gray level co-occurrence matrix were adopted to extract the characteristic parameters of disease color and texture,the preferred characteristic parameters were input into the support vector machine for identification,and the identification results were fed back to the client.Testing experiment showed that the system could accurately and efficiently identify disease types,and enjoy strong practicality as well as good popularization and application prospects.
出处
《安徽大学学报（自然科学版）》
CAS
北大核心
2016年第2期26-31,共6页
Journal of Anhui University(Natural Science Edition)
基金
国家自然科学基金资助项目(61302118)
河南省高校青年骨干教师资助计划(2010GGJS-114)
