Convolutional Virtual Electric Field for Image Segmentation Using Active Contours

• Published in: PloS one Vol. 9; no. 10; p. e110032
• Main Authors: Wang, Yuanquan, Zhu, Ce, Zhang, Jiawan, Jian, Yuden
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 31.10.2014; Public Library of Science (PLoS)
• Subjects: Analysis; Computer and Information Sciences; Computer science; Concavity; Contours; Convergence; Convolution; Electric fields; Electricity; Engineering and Technology; Fast Fourier transformations; Fourier transforms; Humans; Image processing; Image Processing, Computer-Assisted - methods; Image segmentation; Lung - diagnostic imaging; Models, Theoretical; Neural networks; Noise; Pattern recognition; Real time; Segmentation; Signal-To-Noise Ratio; Time Factors; Tomography, X-Ray Computed; User-Computer Interface; Beijing China; China
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0110032

Gradient vector flow (GVF) is an effective external force for active contours; however, it suffers from heavy computation load. The virtual electric field (VEF) model, which can be implemented in real time using fast Fourier transform (FFT), has been proposed later as a remedy for the GVF model. In this work, we present an extension of the VEF model, which is referred to as CONvolutional Virtual Electric Field, CONVEF for short. This proposed CONVEF model takes the VEF model as a convolution operation and employs a modified distance in the convolution kernel. The CONVEF model is also closely related to the vector field convolution (VFC) model. Compared with the GVF, VEF and VFC models, the CONVEF model possesses not only some desirable properties of these models, such as enlarged capture range, u-shape concavity convergence, subject contour convergence and initialization insensitivity, but also some other interesting properties such as G-shape concavity convergence, neighboring objects separation, and noise suppression and simultaneously weak edge preserving. Meanwhile, the CONVEF model can also be implemented in real-time by using FFT. Experimental results illustrate these advantages of the CONVEF model on both synthetic and natural images.