A Segmentation and Extraction Method for Crops of Image in Complex Background

In the traditional fuzzy C-means clustering algorithm (FCM), determination of initial cluster centers and the number of clusters is hard, and interference exists during image segmentation for space information of image isnpsilat considered. In order to overcome the shortcomings above, initial cluster centers and the number of clusters were ascertained by ant colony algorithm, then combine the improved FCM algorithm that considering space information to reduce the image noise. The segmentation and extraction of crops in the image under the complex background was completed. The image that the interference still existed after segmentation would be disposed by the method of mathematical morphology filter. The background was removed completely and the segmentation effect was improved.