マイクロマニプレータの制御 : 第5報, 任意方向を向き多セグメントに欠損を有する生物画像の推論認識法

概要

論文の詳細を見る
In this paper, we propose a reasoning and recognition method for distorted and multidirection organism images for application in a micromanipulator system. In the second report, we proposed an approach for the recognition and identification of microorganisms by using image processing based on the expert system. However, the system could not recognize and identify organism images sufficiently for distorted images. To solve this problem, it is shown here that the system has been improved and can now recognize and identify such images using segment group data and segment connection data. Details of the reasoning and recognition processes of this method are shown in this paper. The effectiveness of this method is proven with experimental results. For the experiments, Aquamicrobes and their two-dimensional black and white images are used.
一般社団法人日本機械学会の論文
1991-04-25