An Automatic Counting System for Oviposited Eggs Based on Digital Image Processing

摘要

In this paper, a new method is described that enables an automatic count by caputering and processing the digital image of the oviposited eggs. The hardware part includes a CCTV camera which hooks up to a personal computer. Oviposited eggs image at the special inspecting stand is captured and stored for future counting process. For automatic counting, our images of oviposited eggs undergo three-stage image processing: Firstly, by applying the fuzzy c-means (FCM) algorithm, oviposited eggs image can be segmented to background and object, then we can get the corresponding fuzzy membership matrix. Every element of this membership matrix falls in the interval [0,255] and represents the strength of association between every pixel and object. Secondly, overlapping oviposited eggs images are separated by the use of the fuzzy distance tranform (FDT) and the watershed algorithm. Thirdly, oviposited eggs are identified and counted. Our experimental results indicate that the proposed method greatly improves the accuracy and efficiency of the oviposited egg counting and the counting result dose not be affected by the subjective factor. Therefore the method can well replace the traditional human method.