OBJECT DETECTION AND CLASSIFICATION THROUGH DEEP LEARNING APPROACHES
Deepika Solanki, Shrwan Ram · Journal of Emerging Technologies and Innovative Research · 2018
In this paper, we implemented the image classification and object detection. This paper presents a deep learning approach for traffic light detection in adapting a single shot detection(SSD) approach and image classification of two categories of bicycle by retraining inceptionv3 model both using an open source tool called TensorFlow Object Detection API. We reviewed the current literature on convolutional object detection and tested the implementability of one of the methods and discovered that convolutional object detection is still evolving as a technology despite that convolutional object detection has outranked other object detection methods. To implement object detection and image classification there is free availability of datasets and pre-trained networks it is possible to create a functional implementation of a deep neural network without access to specialist hardware.