Liquid Medium And Small Target Detection Algorithm Research And Applications

Targets that are smaller than 20 pixels are called small target. The small spatial extend of these targets precludes the use of information about size, shape and other features of a target in traditional image processing techniques. All the research work concerning small target detection are based on two characteristics of small target: the gray level of small targets are different from that of background; the continuity of target track.Detection and tracking of low observable small-targets from a sequence of images against structural background and non-stationary clutter is an important research area in image processing field. Applications of this technique are generally limited to infrared images, especially in Infrared Search and Track (IRST) surveillance applications, which is characteristic by large, low frequency areas and moving clutter. This paper discussed the application of small target detection technique in industry. A series of new algorithms for detecting impurity-one kind of small target in liquid by analyzing sequences of gray scale images were proposed, which are effective for complex structured visible images.Based on the analysis of existing methods and differences between visible images and infrared images, the whole detection process is divided into three main stages: image pre-processing, potential targets extraction and target tracking. Each of them is explained in detail in this paper and the experimental results are presented. In the second stage, those algorithms that are effective for liquid visible image are discussed and several new methods of extracting features of small targets in visible image sequences are proposed. And performances of those algorithms are compared. In the third stage, several target detection and noise repression methods are proposed based on the continuity of target track. Tests of system that integrated effective algorithms on available database have yielded good results.