CAJ | 학술논문

利用图像处理和图像识别技术，对中药叶片图像进行在线识别。主要过程包括图像采集、图像处理、特征提取、图像识别，通过动态规划算法与人工神经网络算法相结合的方法实现叶片图像的识别，平均准确率可达到80%以上。系统基于Web开发，后台采用Java语言实现图像的处理和识别，SQL Server2000数据库实现叶片图像数据的存储。实验结果表明，该系统对中药叶片图像识别具有较好的效果，并且便于操作和实现。
이용도상처리화도상식별기술，대중약협편도상진행재선식별。주요과정포괄도상채집、도상처리、특정제취、도상식별，통과동태규화산법여인공신경망락산법상결합적방법실현협편도상적식별，평균준학솔가체도80%이상。계통기우Web개발，후태채용Java어언실현도상적처리화식별，SQL Server2000수거고실현협편도상수거적존저。실험결과표명，해계통대중약협편도상식별구유교호적효과，병차편우조작화실현。
The on-line identification of TCM leaf image based on image processing and image recognition technology. The main process of it includes image acquisition, image processing, feature extraction, image recognition. Dynamic programming alg-orithm can be combined with artificial neural network algorithm to realize the recognition of leaf image which average accu-racy rate could reach more than 80%.The development of system based on Web;information of all the leaf images can be stored in a SQL Server2000 database. The experimental results show that the system has a good effect on the image recognition of TCM leaf, and facilitate its operation and implementation.