Open AccessShadow Detection and Removal Technique using CNN
Author(s) -
Akanksha Bankhele
Publication year - 2021
Publication title -
international journal for research in applied science and engineering technology
Language(s) - English
Resource type - Journals
ISSN - 2321-9653
DOI - 10.22214/ijraset.2021.37667
Subject(s) - shadow (psychology) , computer science , computer vision , artificial intelligence , convolutional neural network , object detection , orientation (vector space) , image (mathematics) , pattern recognition (psychology) , mathematics , psychology , geometry , psychotherapist
Abstract: The Shadow detection and removal Technique is used in many real-world applications, such as surveillance systems, computer vision applications and indoor outdoor system. The shape and orientation of an object, as well as the light source, can be revealed by shadows in an image. In a traffic surveillance system, the shadow can misclassify the actual target, lowering the system’s accuracy. Numerous algorithms and techniques have been developed by researchers to aid in the detection and removal of shadows in images. This paper aims to provide an overview of different shadow detection and removal techniques, their advantages and drawbacks. Also implementation of Convolutional Neural Network for shadow detection and OpenCV features to remove shadows by re-designing the output and analysing different loss functions to train the network. Keywords: Shadow Detection and Removal Techniques, Shadow Image Processing.