Studies On The Application Of Lawn Conserve Using Digital Image Processing Technology

The application of digital image processing & machine vision technology is an important task on agriculture automation domain. Currently, most researches on lawn images'based on machine vision still have a long way to go in our country. This thesis deeply investigates noise reduction, feature extraction and classification & reorganization enclosing the application of lawn conserve using digital image processing technology. The main contents are listed below:The first, original image is changed from color space to gray, execute noise reduction and the image is processed by gray level transformations. Meanwhile we introduce several median filters and give out their performance.The second, based on characters, we use Hu's moments (seven) and textures based on density intergrowth matrix to describe the lawn images.The third, according to 2 pairs of characters to train improved BP neural net. Finally, this method could classify and recognize the lawn images well.