TABLE OF CONTENT l l Introduction Image Features

TABLE OF CONTENT l l Introduction Image Features 。Color 。Texture Quadtree Decomposition Representative feature extraction 。Vector quantization (VQ) algorithm 。 Representative feature extraction of objects using VQ l Implementation

Introduction Content-Based Image Retrieval l An application of computer vision to the image retrieval problem, that is, the problem of searching for digital images in large databases

Image Features - Color l HSV 。Hue distinguish colors 。Saturation the percentage of white light that is added to a pure color l 。Value perceived light intensity The HSV color model, which is similar to human perception, is most frequently used for retrieval.

Image Features - Color l RGB (Red Green Blue)

Image Features - Texture Angular Second Moment = Contrast = Correlation = Variance = Entropy =

Quadtree Decomposition l l 7 An image is divided into 7 rectangular blocks 7 x 7 is the smallest block we define

Representative feature extraction Vector Quantization (VQ) algorithm l主要概念：配合前述的Quadtree，把圖片重新分割為block 再把切割出來的block配合之前的Texture公式對應到一個8 -dimension vector space中的座標再配合Vector Quantization把vector分群

Representative feature extraction of objects using VQ

Implementation A Simple Follow Chart

Implementation Search Result