|
|
Bimonthly Since 1986 |
ISSN 1004-9037
|
|
|
|
|
Publication Details |
Edited by: Editorial Board of Journal of Data Acquisition and Processing
P.O. Box 2704, Beijing 100190, P.R. China
Sponsored by: Institute of Computing Technology, CAS & China Computer Federation
Undertaken by: Institute of Computing Technology, CAS
Published by: SCIENCE PRESS, BEIJING, CHINA
Distributed by:
China: All Local Post Offices
|
|
|
|
|
|
|
|
|
|
Abstract
Content Based Medical Image Retrieval (CBMIR) is a technique that search relevant images in large repositories according to the content of the query image which is based on selected features such as color, texture, shape or any other features that can be derived from images. The feature extraction and similarity measurement are two vital steps in image retrieval process. The most difficult issue is the semantic gap which the information in the image is lost due to the representation of image in terms of its features. The semantic gap is a significant gap between the representation of image features and visual understanding. Deep learning algorithms produces better results for various machine learning and computer vision. The Deep Learning (DL) possesses various machine learning algorithms. DL methods have been used in many applications areas and these approaches can also be applied in CBMIR. Machine learning techniques can be explored to address the problem of semantic gap that exists. Here we try to reduce the semantic gap by learning discriminative features directly from images by using machine learning techniques and thereby we classify the images. However, deep learning lack generalization and suffer from over fitting whenever trained on small datasets. In order to the obtain better results, these algorithms need larger dataset for the training of the model.
Keyword
Content Based Medical Image Retrieval (CBIR), Artificial Intelligence (AI), Machine Learning (ML) algorithm, query image.
PDF Download (click here)
|
|
|
|
|