Deep Convolutional Level Set Method for Image Segmentation
Journal Title: Journal of ICT Research and Applications - Year 2017, Vol 11, Issue 3
Abstract
Level Set Method is a popular method for image segmentation. One of the problems in Level Set Method is finding the right initial surface parameter, which implicitly affects the curve evolution and ultimately the segmentation result. By setting the initial curve too far away from the target object, Level Set Method could potentially miss the target altogether, whereas by setting the initial curve as general as possible – i.e. capturing the whole image – makes Level Set Method susceptible to noise. Recently, deep-learning methods, especially Convolutional Neural Network (CNN), have been proven to achieve state-of-the-art performance in many computer vision tasks such as image classification and detection. In this paper, a new method is proposed, called Deep Convolutional Level Set Method (DCLSM). The idea is to use the CNN object detector as a prior for Level Set Method segmentation. Using DCLSM it is possible to significantly improve the segmentation accuracy and precision of the classic Level Set Method. It was also found that the prior used in the proposed method is the lower and upper bound for DCLSM’s precision and recall, respectively.
Authors and Affiliations
Agustinus Kristiadi, Pranowo Pranowo
Keywords
Related Articles
- EP ID EP326329
- DOI 10.5614/itbj.ict.res.appl.2017.11.3.5
- Views 125
- Downloads 0
Topic Modeling in Sentiment Analysis: A Systematic Review
With the expansion and acceptance of Word Wide Web, sentiment analysis has become progressively popular research area in information retrieval and web data analysis. Due to the huge amount of user-generated contents over...
An Energy Aware Unequal Clustering Algorithm using Fuzzy Logic for Wireless Sensor Networks
In wireless sensor networks, clustering provides an effective way of organising the sensor nodes to achieve load balancing and increasing the lifetime of the network. Unequal clustering is an extension of common clusteri...
Parallel Technique for Medicinal Plant Identification System using Fuzzy Local Binary Pattern
As biological image databases are growing rapidly, automated species identification based on digital data becomes of great interest for accelerating biodiversity assessment, research and monitoring. This research applied...
Performance Improvement of LeastSquares Adaptive Filter for High-Speed Train Communication Systems
The downlink communication channel from high-altitude platform (HAP) to high-speed train (HST) in the Ka-band is a slowly time-varying Rician distributed flat fading channel with 10-25 dB Rician K factor. In this respect...
VLSI Architecture for Configurable and Low-Complexity Design of Hard-Decision Viterbi Decoding Algorithm
Convolutional encoding and data decoding are fundamental processes in convolutional error correction. One of the most popular error correction methods in decoding is the Viterbi algorithm. It is extensively implemented i...