Video 33: Content Based Image Retrieval (CBIR) using Wavelet Features, CLD and EHD of MPEG-7

Hello Viewers, in this video, a Content Based Image Retrieval (CBIR) system is implemented. This CBIR utilizes Color Layout Descriptor (CLD) for color feature extraction and Wavelets and Edge Histogram Descriptor (EHD) for extraction of texture features.

This CBIR uses a feature vector of just length of 181 and it also shows enough robustness against distorted input images.

The implementation is done in MATLAB.

This video has following contents:

  • What is Content Based Image Retrieval (CBIR)? 
  • CLD and EHD as feature vectors.
  • Use of Wavelets for features extraction.
  • Image Databases used for image retrieval.
  • Training Procedure and Testing Procedure.
  • Performance Parameters for a CBIR system.
  • MATLAB code of complete CBIR system.

Link of previous videos:

1. Color Layout Descriptor (CLD) of Mpeg-7 for Image Retrieval: Click Here

2. Edge Histogram Descriptor (EHD) of Mpeg-7 for Image Retrieval: Click Here

Link of Image Databases:

1. Wang Image Database:​

2. Microsoft Image Database:

Download Resources: 

1. Test Images: Download

2. Wang Database: Download

3. Microsoft Research Database (Reduced): Download


Post a Comment