CROP DETECTION BY MACHINE VISION FOR WEED MANAGEMENT

Weed management is one of the costliest input to the agriculture and it is one of the un-mechanised area. To bring mechanization in this area the most important step is the detection of weed in agricultural field. Weed can be detected by using machine vision techniques. Machine vision uses special image processing techniques. Weeds in agricultural field can be detected by its properties such as Size, Shape, Spectral Reflectance, Texture features. In this paper we are demonstrating weed detection by its Size features. After the image acquisition Excessive green algorithm is developed to remove soil and other unnecessary objects from the image. Image enhancement techniques are used to remove Noise from the images, By using Labelling algorithm each components in the Image were extracted, then size based features like Area, Perimeter, longest chord and longest perpendicular chord are calculated for each label and by selecting appropriate threshold value Weed and Crop segmentation is done . Result of all features is compared to get the best result.