Informatics Publishing Limited

Ashima Mehta ¹, Anuj Kumar Sharma ¹, Puneet ²

Affiliations
1 Department of CSE, OITM, Hisar, Haryana, IN
2 Chandigarh University, Mohali, Punjab, IN

Abstract

Color is one of the most imperative low-level features used in image retrieval and most content-based image retrievals (CBIR) schemes use color as an image features. However, imagere possession using only color features often provide very inadequate outcomes because in many cases, images with comparable colors do not have similar content. As the clarification of thisproblem this paper describes a novel algorithm for Content Based Image Retrieval (CBIR) based on Color Edge Detection and Discrete Wavelet Transform (DWT). This method is different from the existing histogram based techniques. Theproposed algorihm generates feature vectors that conglomerate both color and edge features. This documents also uses wavelet transform to diminish the size of the feature vector and concurrently conserving the content details. The forcefulness ofthe structure is also tested against query image adjustments such as geometric distortions and noise calculation etc. Wang's image database is used for tentative study and results are presented in terms of precision and recall.

Keywords

YCbCr, RGB, Canny Edge Detection, Color Edge Detection, Wavelet Transform.

Full Text

     

Ashima Mehta ¹, Anuj Kumar Sharma ¹

Affiliations
1 Department of CSE, OITM, Hisar, Haryana, IN

Abstract

This paper show cases a novel procedures for detecting faces in color images using combine with skin color segmentation. First, skin color model in the YCbCr chrominance space is built to segment the non-skin-color pixels from the image. Then, mathematical morphological operators are used to remove noisy regions and fill holes in the skin-color region, so we can extract applicant human face regions. Finally, these face applicants are scanned by Cascade classifier based on AdaBoost for more meticulous face detection. This system perceives human face in different scales, various postures, different utterance, lighting conditions, and assimilation. Exploratory results show the proposed system obtains competitive results and improves detection appearance significantly.

Keywords

Face Recognition, Color Images, Skin Color.

Full Text