Image data field-based framework for image thresholding

Thresholding is a popular image segmentation method that converts a gray level image into a binary image. In this paper, we propose an image data field-based framework for image thresholding, and improve four selected methods under the proposed framework, which involves three major steps, generating the image data field, implementing image transformation with potential-weighted sum, and then determining the binary threshold for the transformed image by applying the conventional approaches. Image data field keeps the balance between spatial and grayscale information in local neighbourhood by potential calculation, and keeps the balance between the local information and the global trend by image data field generator. Compared with the original algorithms on a variety of synthetic and real images, with or without noise, the experimental results suggest that the presented method is accurate, noise-robust, efficient and scalable. (C) 2014 Elsevier Ltd. All rights reserved.