An IoT based modified graph cut segmentation with optimized adaptive connectivity and shape priors

Abstract Internet of Things (IoT) is the network of physical objects embedded with software, electronics, sensors and network connectivity which enables these objects to collect and exchange data. IoT based image processing is reaching up to a different level which is becoming a major part of everyone life. IoT based image segmentation is used to solve the real world problems like animal classification in wildlife monitoring and vehicle classification etc. Image segmentation is one of the processes of image processing technique which involves partitioning an image into multiple segments so that any region of interest can be obtained easily. The modified graph cut approach is one of the image segmentation techniques that avoid the boundary and region balance problem of graph cut technique by removing the regional term and strengthening the boundary term with the prior information. By combining shape priors in an adaptive way, a robust way to harness the shape prior was introduced in graphcut segmentation that reduced the memory size required for modified graph cut implementation. However, the structural regularities were not handled by the modified graph cut with adaptive shape prior method. Hence, in this proposed work the structural irregularities in image shapes are considered for image segmentation where the connectivity constraints are formulated such that corresponding optimization problems are all NP-hard which is solved by optimized DijkstraGC and problem decomposition approach. The segmentation with satisfied constraint results in an efficient image segmentation which reduces amount of time required for segmentation process. Furthermore, parameter of adaptive shape prior and connectivity constraints are optimized using BAT algorithm to improve the performance of image segmentation. Results clearly shows that proposed approach gives better results in terms of Root Mean Square, Energy Consumption and correlation coefficient.

[1]  N. Jamil,et al.  An accurate thresholding-based segmentation technique for natural images , 2012, 2012 IEEE Symposium on Humanities, Science and Engineering Research.

[2]  Wenbing Tao,et al.  Multilayer graph cuts based unsupervised color-texture image segmentation using multivariate mixed student's t-distribution and regional credibility merging , 2013, Pattern Recognit..

[3]  Su Ruan,et al.  Multilabel statistical shape prior for image segmentation , 2016, IET Image Process..

[4]  Piyush Vyas,et al.  A robust technique for image processing based on interfacing of Raspberry-Pi and FPGA using IoT , 2017, 2017 International Conference on Computer, Communications and Electronics (Comptelix).

[5]  Tao Wang,et al.  Image segmentation based on weighting boundary information via graph cut , 2015, J. Vis. Commun. Image Represent..

[6]  Weixing Wang,et al.  Efficient multilevel image segmentation through fuzzy entropy maximization and graph cut optimization , 2014, Pattern Recognit..

[7]  Hao Gao,et al.  Multilevel Thresholding for Image Segmentation Through an Improved Quantum-Behaved Particle Swarm Algorithm , 2010, IEEE Transactions on Instrumentation and Measurement.

[8]  Sung-Jea Ko,et al.  Image segmentation based on modified graph-cut algorithm , 2010 .

[9]  Ezzeddine Zagrouba,et al.  Model-based graph-cut method for automatic flower segmentation with spatial constraints , 2014, Image Vis. Comput..

[10]  Eros Comunello,et al.  Learning a color distance metric for region-based image segmentation , 2009, Pattern Recognit. Lett..

[11]  Khalid M. Amin,et al.  A new Optimization-Based Image Segmentation method By Particle Swarm Optimization , 2011 .

[12]  Michael Unser,et al.  Efficient Shape Priors for Spline-Based Snakes , 2015, IEEE Transactions on Image Processing.

[13]  J. Dinesh Peter,et al.  Modified Graph-Cut Algorithm with Adaptive Shape Prior , 2013, ICACNI.

[14]  Dwarikanath Mahapatra,et al.  Combining multiple expert annotations using semi-supervised learning and graph cuts for medical image segmentation , 2016, Comput. Vis. Image Underst..

[15]  Yu Li,et al.  A fuzzy clustering image segmentation algorithm based on Hidden Markov Random Field models and Voronoi Tessellation , 2017, Pattern Recognit. Lett..

[16]  Vijai Singh,et al.  Reboost Image Segmentation using Genetic Algorithm , 2013 .

[17]  Marcelo Pereyra,et al.  Fast Unsupervised Bayesian Image Segmentation With Adaptive Spatial Regularisation , 2015, IEEE Transactions on Image Processing.

[18]  M Radha,et al.  EDGE DETECTION TECHNIQUES FOR IMAGE SEGMENTATION , 2011 .

[19]  Hui Wang,et al.  Adaptive shape prior in graph cut image segmentation , 2013, Pattern Recognit..

[20]  Yi Gao,et al.  Automatic Segmentation of the Left Atrium From MR Images via Variational Region Growing With a Moments-Based Shape Prior , 2013, IEEE Transactions on Image Processing.

[21]  Sushila Shidnal,et al.  Implementation of IoT (Internet of Things) and Image processing in smart agriculture , 2016, 2016 International Conference on Computation System and Information Technology for Sustainable Solutions (CSITSS).

[22]  Amritpal Kaur,et al.  Image Segmentation with Artificial Neural Networs Alongwith Updated Jseg Algorithm , 2014 .

[23]  Andrew Zisserman,et al.  OBJCUT: Efficient Segmentation Using Top-Down and Bottom-Up Cues , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.