Scribd
Upload
38 views

Mid Semester Project Seminar On: Content Based Image Retrieval

The document summarizes a mid-semester project seminar on content-based image retrieval. It discusses extracting features like shape, color, and texture to build image vectors for querying image databases. It proposes implementing extraction of image vectors and primary comparisons based on Euclidean distance for a desktop application in Phase I, and adding human factors and a web interface in Phase II. Key features are shape recognition using edge detection and form matrices, and representing color using spatial chromatic histograms and texture using texture signatures.

Uploaded by

Sidharth Aggarwal
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
38 views

Mid Semester Project Seminar On: Content Based Image Retrieval

The document summarizes a mid-semester project seminar on content-based image retrieval. It discusses extracting features like shape, color, and texture to build image vectors for querying image databases. It proposes implementing extraction of image vectors and primary comparisons based on Euclidean distance for a desktop application in Phase I, and adding human factors and a web interface in Phase II. Key features are shape recognition using edge detection and form matrices, and representing color using spatial chromatic histograms and texture using texture signatures.

Uploaded by

Sidharth Aggarwal
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 21

Mid Semester Project Seminar on Content Based Image Retrieval

Submitted By: Rachit Kumar(8103410) Kamesh Sharma(8104687) Sidharth Aggarwal(8104698)

Introduction
The recent tremendous growth in computer technology has also brought a substantial increase in the storage of digital imagery. Storage of such image data is relatively straightforward, but accessing and searching image databases is intrinsically harder than their textual counterparts. The goal of Content-Based Image Retrieval (CBIR) systems is to operate on collections of images and, in response to visual queries, extract relevant image.

Significance

Medical Image Databases. Trademark Image Registration. Automatic face recognition systems. Search for one specific image.

General browsing to make an interactive choice.

Two Classes of CBIR Narrow vs. Broad Domain


Narrow
Medical Imagery Retrieval Finger Print Retrieval Satellite Imagery Retrieval Broad

Photo Collections
Internet

Our Domain
We aim for Broad Perspective.

Phase I (7th Semester) : Implement extraction of image vectors and primary comparisons based on Euclidean distance. Implement for Desktop based application.
Phase II (8th Semester) : Implement the results with stress on Human Relations and go for Web Domain.

Our Approach (SCT)


Features Focused
Shape Color

Texture

SHAPE

Overview(Shape)
Image Shape Recognitions Edge Detections Form Vector Matrices

Scale the Images

Overview(Shape)
Recognize approximate images Choosing Prototypes Region Grouping

Overview(Shape)
Regularization of Scaling Behavior. The Empirical Robustness Evaluation (ERV)

Color

Overview(Color)
Color-based image retrieval using Spatial-Chromatic Histogram
Information about the location of pixel having the same color and their arrangement within the image By SCH(Spatial Chromatic Histogram) has been defined to answer the following question

1. Where are the pixel with similar color approximately located?


2. How are these pixel spatially arranged?

Overview(Color)
Content Bases Image Retrieval Using Color , Texture and Shape
An integrated matching scheme ,based on Most Similar Highest Priority(MSHP) principle. The adjacency matrix of a bipartite graph formed using the tiles of query and target image. The significance of a region plays an important Role in the image matching process.

Overview(Color)
Image Retrieval based on Fuzzy Color Histogram
Advantage : It is fast to manipulate, store and compare and incentive to Rotation and Scale Disadvantage : It is also quite unreliable or it is sensitive to even a small change in the Scene of the image. A histogram is created by firstly dividing a color space into a number of bins and then by counting the number of pixels of the image that belongs to each bin

TEXTURE

Overview(Texture)
Texture is some visual patters defined in terms of Granularity, Directionality, and Repetitiveness. Different approaches to extract and represent textures -space base descriptor -frequency based descriptor

-texture signature

Overview(Texture)
The Co-occurrence Matrix: Based on the orientation and distance between image pixels from it we obtain statistics that represent (CCRDR) : *Coarseness
*Contrast *Regularity *Directionality *Roughness

Overview(Texture)
Frequency based texture descriptors:

Garbor Wavelet coefficientsThe wavelet transform transforms the image into a multiscale representation with both spatial and frequency characteristics. Texture Histogram Descriptor -Histogram of images are computed and entropy, smoothness and uniformly ,mean ,second moment, third moment and fourth moment. -comparison of images based on the H.S.V channel of the image

Overview(Texture)
Retrieval efficiency The Common Evaluation Measures used in CBIR Systems

The Precision = No. of relevant images retrieved/total no. of images retrieved The Recall = No. of relevant images retrieved /total no. of relevant images in the database

References
Shape Matching and Object Recognition Using Shape Context (Serge Belongie, Jitendra Malik and Jan Puzicha) Web Link Content Based Image Retrieval Theory and Application(Ricardo da Silva Torres Alexandre Xavior Falcao) Web Link Shape Matching by Region Grouping(Yan Gao WeiYun Yau) Web Link

References
Color Based Image Retrieval using SpatialChromatic Histogram (L.Cinque,G. Ciocca, S.Levialdi, A.Pellicano, R.Schettini) Web Link Content based Image Retrival using Color Texture and Shape(P.SHiremath, Jagadeesh Pujari) Web Link Image Retrival Based on Fuzzy Color Histogram(Kaixing wu,Qiang Xu) Web Link

You might also like