Loading [MathJax]/extensions/MathZoom.js
Super-pixel Segmentation based Skin texture pattern recognition | IEEE Conference Publication | IEEE Xplore

Super-pixel Segmentation based Skin texture pattern recognition


Abstract:

Super-pixel segmentation is widely used nowadays in image processing to enhance segmentation accuracy. A new detection model is proposed for skin texture pattern recognit...Show More

Abstract:

Super-pixel segmentation is widely used nowadays in image processing to enhance segmentation accuracy. A new detection model is proposed for skin texture pattern recognition of Leopard, Cheetah, and Jaguar. In this model, a combination of Histogram of Gradient (HOG) and superpixel segmentation is used for extracting the features and the segmentation task of the target animal. This method does not require several superpixels to be created in advance, whereas it can automatically partition the image to its content into a suitable number of superpixels without any over or under segmentation. Then, the obtained features are fed into a Support Vector Machine (SVM) classifier to classify the skin texture pattern of Leopard, Cheetah, and Jaguar. The validation is performed which shows that the classifier achieves an accuracy of 96.67 %.
Date of Conference: 02-04 December 2021
Date Added to IEEE Xplore: 20 January 2022
ISBN Information:
Conference Location: Coimbatore, India

Funding Agency:


I. Introduction

Nowadays automated computer vision application helps in identification of individual animal with its skin texture pattern. It's easy to differentiate different animals with their skin pattern. Biological studies and conservation efforts have made a manual recognition of individual animals with their shape and skin pattern. Texture and pattern play an important role in understanding and description of an object. Conservation research scientists deal with skin texture to recognize wild animals. It would be very helpful for the scientist with a system that can automatically identify animal skin texture patterns for the classification of wildlife with camera trap images. Some of the challenges in camera trap images are variations in point of view, occluded image, the dataset for testing, and training texture for classification. In this paper, samples of Leopard, Cheetah, and Jaguar are collected for skin texture pattern recognition. The skin texture pattern of Cheetah has a black solid spot throughout its body, Leopard has a rosette pattern and Jaguar have a rosette with a black solid spot in the center. Pattern recognition is the identification of patterns by employing a machine learning algorithm It discriminates the input data based on the information gathered from the patterns of input data. Based on the recognition of skin pattern the classifier will give its decision whether it is Cheetah, Jaguar, or Leopard. There are many algorithms for pattern recognition such as Neural Network, Template matching, Fuzzy model, etc.

References

References is not available for this document.