In order to obtain the timely spatial distribution and the accurate area of crops efficiently, a detection method for field crop was established based on unmanned aerial vehicle(UAV) platform and image analysis. A small four rotor UAV equipped with high performance digital camera was adopted to get the crop image. The target samples were obtained through primitive segmentation and visual interpretation, to extract 21 dimensional color features and 3 dimensional texture features. BP neural network classifier and pixel accumulation method were used to identify the crop species and measure respective area. Test results show that the average recognition rate of wheat, rapeseeds, broad beans and garlic was achieved to be 86% with the average relative error of 9.62% for area measurement.