High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Interactive Object Identification using Image Recognition


Jayanthi , R.M.K Engineering College; Harshini. S, R.M.K Engineering College; Duggi Meghana, R.M.K Engineering College


Image Processing; Neural Networks; Machine Learning; Natural Language Processing; Computer Vision; Convolution Neural Network; Recurrent Neural Network; Deep Learning; Long Short Term Memory Units


This project is proposed to help visually challenged people. It manipulates to image processing and natural language processing techniques to simulate human vision. The proposed system takes images and other multimedia files and forms logical sequence between them and help the users understand it. We focus on recognizing human actions in still images, which is done by analyzing human poses and their interaction with objects. A systematic approach of recognizing objects and their relationship is developed through this project. In the image processing technology, automatically generating a natural language description of an image is an important task. A multi-model neural network system is used in this paper which describes the content of images automatically. This multi-model neural network is divided into an object detection and localization model, which extract the information of objects and their spatial relationship in images respectively. Sentences generation is done by using long short-term memory (LSTM) units with attention mechanism in a deep recurrent neural network (RNN). This can also be used to provide highly sophisticated search and in forensic systems since mining data from multimedia files greatly improves the search and data analytics.

Other Details

Paper ID: IJSRDV6I20690
Published in: Volume : 6, Issue : 2
Publication Date: 01/05/2018
Page(s): 1059-1062

Article Preview

Download Article