UNDER WATER IMAGE USING CONTRAST LIMITED ADAPTIVE HISTOGRAM EQUALIZATION
Author(s):
DR.D. CHITRA, DR.S. SARASWATHI, K. RASIGA
Keywords:
Underwater Imaging, Histogram Equalization & Algorithms
Abstract
This paper describes about the underwater computer vision. Underwater image processing is in general a challenging task, due to its environment, poor sunlight and the turbidity in itself. Optical, sonar and ultrasound images are captured from the underwater environment. Often optical cameras seem to be a good choice, especially in the case of underwater species identification or counting, coral reefs, pipeline monitoring, mining etc. Underwater images captured from such cameras are poor in contrast, blurred and often contain noise due to the flora and fauna floating in the water. The visibility is limited due to the fact that when light enters the water it is exponentially attenuated. In some cases, underwater images are captured in very low illumination such that object detection itself becomes a challenging task. For accurate object recognition, underwater images must undergo initial pre-processing. This preprocessing must include image enhancement and de- noising as the image suffers from poor contrast, non-uniform lighting, blurring etc. The extracted features must undergo an Artificial Intelligence algorithm for classification. Such automation is often required, as ocean floor is monitored continuously for various applications and manual identification will not help. Artificial neural networks, which exhibit capabilities for adaptation, was used for classifying the objects of interest underwater. The learning algorithms were designed such that features were trained and back propagated to the hidden neurons, until the error was minimized. A stepwise feature selection process was used to determine the subset of features that will optimize the probability of detection and classification. This resulted in accurate object recognition with 96% classification accuracy.
Article Details
Unique Paper ID: 151541

Publication Volume & Issue: Volume 8, Issue 1

Page(s): 225 - 235
Article Preview & Download


Share This Article

Conference Alert

ICM - STEP

International conference on Management, Science, Technology, Engineering, Pharmact and Humanities.

Go To Issue



Call For Paper

Volume 7 Issue 9

Last Date 25 February 2020

About Us

IJIRT.org enables door in research by providing high quality research articles in open access market.

Send us any query related to your research on editor@ijirt.org

Social Media

Google Verified Reviews

Contact Details

Telephone:6351679790
Email: editor@ijirt.org
Website: ijirt.org

Policies