Undersea Water Objects Detection and Classification Using Machine Learning and Deep Learning Techniques

Loading...
Thumbnail Image

Date

item.page.authors

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

In our daily lives, the object detection task is widely used in various applications such as medical imaging systems, pedestrian location and etc. Exploring this mysterious region of the ocean is becoming more popular as underwater observations and exploitation advance quickly. An underwater image applicability and visual perception improvement are the aim of underwater image enhancement. An original underwater image with its visual appearance is a challenging task and it is affected due to low image contrast, blurring, more noises and etc. Digital cameras are employed to record images of the ocean floor in which the most of images are noisy, low contrast and uneven illumination characteristics due to absorption and scattering of light effect in the water. Hence, image processing technique is frequently used to improve the quality and color fading as well as reduce noise. For object detection, the most significant procedure is image preprocessing of underwater images. Because of the effects of light scattering and absorption in the water, the images obtained by the underwater vision system show the characteristics of uneven illumination, low contrast, and serious noise. Contour segmentation techniques are frequently used to identify objects in these images. Most of the studies provided non-appropriate segmentation results of undersea water images. Finally, the detection and classification step to classify the undersea objects from the image. In the existing studies, we have encountered lot of issues in case of computational complexity, minimum classification accuracy, more time execution, less training samples and etc. To achieve desired goal, we have used Gaussian Golden Search Optimization based on Machine Learning Algorithms to detect the undersea water image object detection. Here, we have applied under water image pre-processing, segmentation with object detection and classification using Gaussian Golden Search Optimization based Support Vector Machine model. We have also used optimized Deep Convolutional Ne

Description

Keywords

Citation

item.page.endorsement

item.page.review

item.page.supplemented

item.page.referenced