Most image analysis methods perform segmentation as a first step towards producing the object description. In these methods both input and output are images only, but the output is an abstract representation of the input. Image segmentation is responsible for partitioning an image into multiple sub-regions based on a desired feature, such as edge, point, line, boundary, texture, and region. The most common segmentation methods use intensity-based images. Snakes or active contours (AC) are used extensively in computer vision and image processing applications, particularly to locate object boundaries. During a literature survey, it has been identified that many segmentation issues like gray values of pixels remain equal for a area, rapid change of image gradient, large gradient due to noise and identification of boundary between areas. The traditional adaptive distance preserving level set evolution method works fine for natural and synthetic images. However, it is necessary to set the performance parameters manually every time with respect to the type of input image. Hence, the proposed method of an automatic parameter setting model (APSM) improves the adaptive distance preserving level set evolution based on region by setting the performance parameters automatically with the help of an analysis of the image quality. The results show the effective performance of segmentation by reducing the number of iterations with an improved output quality.
CITATION STYLE
Baswaraj, D., & Prasad, P. S. (2019). Efficient image segmentation using an automatic parameter setting model. In Lecture Notes in Electrical Engineering (Vol. 500, pp. 529–542). Springer Verlag. https://doi.org/10.1007/978-981-13-0212-1_55
Mendeley helps you to discover research relevant for your work.