Abstract: Crop-row detection is a dilemma, which was widely studied in the world. In this article, research on crop-row detection was reviewed and the characteristics of field crop planting and its early growth information were described. To detect the position of crop-rows, a method based on Hough transform and connected component labeling was put forward. First, a true color image of field crops was processed with gray histogram and OSTU respectively. Then the noises in the image were eliminated by means of eight connected component labeling according to their areas. The centroid coordinates of each labeled crop were computed after the second labeling, through which the center line of the crop-row was regressed by Hough transform. Experimental results show that the method is effective.