Animal detection based on deep convolutional neural networks with genetic segmentation

This paper presents a system for automatic detection and recognition of the animals using Deep CNN with genetic segmentation. In the present work, the grouping of input animal pictures is done with the help of a Convolutional Neural Network is demonstrated. The proposed work is compared with standar...

Full description

Saved in:
Bibliographic Details
Published in:Multimedia tools and applications 2022-12, Vol.81 (29), p.42149-42162
Main Authors: Chandrakar, Ramakant, Raja, Rohit, Miri, Rohit
Format: Article
Language:English
Subjects:
1220: Visual and Sensory Data Processing for Real Time Intelligent Surveillance System
Artificial neural networks
Computer Communication Networks
Computer Science
Data Structures and Information Theory
Genetic algorithms
Multimedia Information Systems
Neural networks
Pictures
Recall
Recognition
Segmentation
Special Purpose and Application-Based Systems
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper presents a system for automatic detection and recognition of the animals using Deep CNN with genetic segmentation. In the present work, the grouping of input animal pictures is done with the help of a Convolutional Neural Network is demonstrated. The proposed work is compared with standard recognition methods such as SU, DS, MDF, LEGS, DRFI, MR, GC. The existing methodologies have more error rates because of high false-positive & negative rate detection, hence there is a need for a highly accurate system for animal detection. According to the proposed work, a genetic algorithm is used for the segmentation process, and for classification 3-layers neural network is used. For training and examining the proposed work, a database is created which consists of 100 distinct subjects with 2 classes and 10 pictures in each class. Experimental results are demonstrated as the segmentation using genetic algorithms and the novelty of the proposed method in terms of precision, recall, f-measurement, and MAE. Hence proposed work improves the overall results i.e. precision (99.02%), recall (98.79%), F-Measurement (98.9%), and MAE (0.78%).
ISSN:1380-7501
1573-7721
DOI:10.1007/s11042-021-11290-4