The segmentation of rice seedlings by taking multispectral image of paddy field helps to measure the forecasting yield, nutrition in crops, supervising diseases and insect pest, etc. Although it is intricate to perform image segmentation of crop from the background as there are various factors affecting the outcome such as light condition, water reflection, complex environment and background. This research used ‘Difference Vegetation Index – DVI’ image to reduce noise for identifying seedling using the spectral characteristics of the objects in near infrared, red and green wavebands. The high resolution multispectral camera plays an important role in this research. The segmentation was highly accurate which achieved over 99% whereas the segmentation method processed fast and gave well output