科研之友
微信
新浪微博
Facebook
分享链接
摘要
A novel method for image recognition of bacteria is presented in this paper. An effective method for the edge detection of bacteria images is provided by optimising segmentation threshold and changing the order of edge detection, which effectively removes or restrains noise and gets clear edges of the microscopic image. Principal component analysis (PCA) is used to reduce the dimensionality of primitive features extracted from the images, and back propagation (BP) network is used to learn to classify the images. The proposed method was tested with National Microbe Culture Resource (NMCR) database, and experimental results demonstrated that the presented image classification scheme is effective.
