Analysis of Different Feature Description Algorithm in object Recognition

Analysis of Different Feature Description Algorithm in object Recognition

Sirshendu Hore (HETC, India), Sankhadeep Chatterjee (University of Calcutta, India), Shouvik Chakraborty (University of Kalyani, India) and Rahul Kumar Shaw (HETC, India)
Copyright: © 2018 |Pages: 35
DOI: 10.4018/978-1-5225-5204-8.ch023
OnDemand PDF Download:
List Price: $37.50
10% Discount:-$3.75


Object recognition can be done based on local feature description algorithm or through global feature description algorithm. Both types of these descriptors have the efficiency in recognizing an object quickly and accurately. The proposed work judges their performance in different circumstances such as rotational effect scaling effect, illumination effect and blurring effect. Authors also investigate the speed of each algorithm in different situations. The experimental result shows that each one has some advantages as well as some drawbacks. SIFT (Scale Invariant Feature Transformation) and SURF (Speeded Up Robust Features) performs relatively better under scale and rotation change. MSER (Maximally stable extremal regions) performs better under scale change, MinEigen in affine change and illumination change while FAST (Feature from Accelerated segment test) and SURF consume less time.
Chapter Preview


Digital image processing makes the use of different algorithms to accomplish image processing on digital images. The object detection and extraction of feature from the object plays a vital role in case of digital image processing. To obtain some useful information from different digital media such as photo, video or any form of multimedia content digital Image processing relies heavily on feature extraction and object detection and subsequent object recognition. Successful and efficient object recognition is an important research domain in computer vision and image processing. Though object recognition has started it journey four decade back, it has started making its acceptance rapidly in recent years due to the advances in computational intelligence. It is also influenced by the advancement made in the field of feature extraction techniques Object recognition is the process of determining the distinctiveness of an object being perceived in the image. This is often done using a set of known labels. Significant effort has been made earlier to develop some generic made to overcome the challenges often encountered in the case of object recognition. Recognition of object in cognitive way is much easier then recognizing the same object through computer vision or image processing. Pose of an object relative to a camera, variation in lighting under different condition, and difficulty in generalizing across objects from a set of images causes much difficulties in object recognition process. In the literature different way of recognizing an object is reported.

Complete Chapter List

Search this Book: