MSER-in-Chip: An Efficient Vision Tool for IoT Devices

MSERs (maximally stable extremal regions) belong to the most popular local image features with a wide range of highly practical applications, e.g., (to name a few) in image search and retrieval, object detection and recognition, image stitching, tracking mobile objects, etc. Low complexity of MSER detectors (combined with a regular structure suitable for hardware implementation) makes MSER an attractive option for IoT devices which may require vision capabilities to analyze and interpret their environments. In this chapter, we briefly discuss theoretical, algorithmic, and technical aspects of using MSER for such purposes. The presented results have contributed to the system-on-chip implementation of MSER detector which is also overviewed in this chapter. Additionally, we discuss prospective hardware implementations of even more efficient feature detectors based on MSERs.

[1]  Michael H. F. Wilkinson,et al.  A Comparison of Algorithms for Connected Set Openings and Closings , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Horst Bischof,et al.  Color Blob Segmentation by MSER Analysis , 2006, 2006 International Conference on Image Processing.

[3]  James J. Little,et al.  Curious George: An attentive semantic robot , 2008, Robotics Auton. Syst..

[4]  Aamna Alali,et al.  Visual Target Tracking Using a Low-Cost Methodology Based on Visual Words , 2016, ICCVG.

[5]  Andrzej Sluzek,et al.  A maximally stable extremal regions system-on-chip for real-time visual surveillance , 2015, IECON 2015 - 41st Annual Conference of the IEEE Industrial Electronics Society.

[6]  David Nistér,et al.  Linear Time Maximally Stable Extremal Regions , 2008, ECCV.

[7]  Andrew Zisserman,et al.  Video data mining using configurations of viewpoint invariant regions , 2004, CVPR 2004.

[8]  Cordelia Schmid,et al.  A Comparison of Affine Region Detectors , 2005, International Journal of Computer Vision.

[9]  Andrzej Sluzek,et al.  Multi-distinctive MSER Features and Their Descriptors: A Low-Complexity Tool for Image Matching , 2015, ACIVS.

[10]  Horst Bischof,et al.  Detecting, Tracking and Recognizing License Plates , 2007, ACCV.

[11]  Robert C. Bolles,et al.  Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography , 1981, CACM.

[12]  Andrzej Sluzek,et al.  A hardware accelerator for real-time extraction of the linear-time MSER algorithm , 2015, IECON 2015 - 41st Annual Conference of the IEEE Industrial Electronics Society.

[13]  Simone Calderara,et al.  Visual Tracking: An Experimental Survey , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Andrzej Sluzek,et al.  Algorithmic foundations for hardware implementation of scale-insensitive MSER features , 2016, 2016 IEEE 59th International Midwest Symposium on Circuits and Systems (MWSCAS).

[15]  Chong-Wah Ngo,et al.  Keyframe Retrieval by Keypoints: Can Point-to-Point Matching Help? , 2006, CIVR.

[16]  W. James MacLean,et al.  Real-Time Extraction of Maximally Stable Extremal Regions on an FPGA , 2007, 2007 IEEE International Symposium on Circuits and Systems.

[17]  Jiri Matas,et al.  Robust wide-baseline stereo from maximally stable extremal regions , 2004, Image Vis. Comput..

[18]  Alexander M. Bronstein,et al.  Are MSER Features Really Interesting? , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[19]  Vincent Lepetit,et al.  Monocular Model-Based 3D Tracking of Rigid Objects: A Survey , 2005, Found. Trends Comput. Graph. Vis..

[20]  T. Lindeberg.  Scale-Space Theory : A Basic Tool for Analysing Structures at Different Scales , 1994 .

[21]  Andrzej Sluzek,et al.  Improving Performances of MSER Features in Matching and Retrieval Tasks , 2016, ECCV Workshops.