Carp-DCAE: Deep convolutional autoencoder for carp fish classification

•The latent features of Simple (SAE), Deep (DAE), and Convolutional deep (DCAE) autoencoder models are extracted and utilised to identify the major carp species. The experiment is conducted using a standard dataset of 1500 photos (500 images of each species) of three major carp species: Labeo catla...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computers and electronics in agriculture Jg. 196; S. 106810
Hauptverfasser: Banerjee, Arnab, Das, Arijit, Behra, Samarendra, Bhattacharjee, Debotosh, Srinivasan, Nagesh Talagunda, Nasipuri, Mita, Das, Nibaran
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Amsterdam Elsevier B.V 01.05.2022
Elsevier BV
Schlagworte:
agriculture
carp
Classifiers
data collection
Deep learning
electronics
Feature recognition
Fish
Fish Identification
Fisheries
Fully Connected Layers
Image Processing
India
industry
Machine learning
Pattern Recognition
regression analysis
Representations
Segmentation
socioeconomics
species identification
Support vector machines
texture
India
Fish Identification
Pattern Recognition
Fully Connected Layers
Segmentation
Image Processing
ISSN:0168-1699, 1872-7107
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•The latent features of Simple (SAE), Deep (DAE), and Convolutional deep (DCAE) autoencoder models are extracted and utilised to identify the major carp species. The experiment is conducted using a standard dataset of 1500 photos (500 images of each species) of three major carp species: Labeo catla (Catla), Labeo rohita (Rohu), and Cirrhinus cirrhosus (Mrigal).•The major carp fish species are identified using two methods: one involves adding a fully connected layer at the end of the autoencoder network, and the other involves combining the autoencoder network with typical machine learning algorithms such as Logistic Regression, Naive-Bayse, K-Nearest Neighbor, Support Vector Machine, and Random Forest.•All of the experiments are carried out incrementally on the acquired data in order to assess the efficacy of our suggested approach as the amount of data increases.•The results of the proposed approach is compared with the result using some traditional image descriptors: 206 Hu moments, Haralick texture, WLD, and HOG as well as some popular deep neural networks like InceptionV3, InceptionResNetV2, MobileNet, VGG16 and VGG19. The fisheries industry relies heavily on automatic fish species identification for its socio-economic well-being. Due to the similarity in shape and size of the major carps , it can be difficult to recognise them using morphological features. To recognise these species automatically, our proposed autoencoder network models have been applied to a fish dataset containing 1500 images of three major carps of India. As a feature, the autoencoder model’s latent representation is used. After the training phase is complete, the decoder is removed and fish species are categorised using several classifiers. Different variations of autoencoders, such as the Simple Autoencoder, the Deep Autoencoder, and the Deep Convolutional Autoencoder are applied with different hyper parameters. An encouraging maximum accuracy rate of 97.33% is obtained in 250 epochs with a learning rate of 0.0001 using Deep Convolutional Autoencoder. Some well-known machine learning classifiers, such as Logistic Regression, Naive Bayes, K-Nearest Neighbor, Support Vector Machine, and Random Forest, are also used to evaluate the latent representation’s effectiveness using the latent representation as a feature vector. The Support Vector Machine-based latent representation of the Deep Convolutional Autoencoder outperformed all other approaches significantly. The models’ performance is compared to that of Hu moments, Haralick texture, Weber local descriptor, HOG descriptor etc. with best classifiers along with different deep learning models, such as InceptionV3, InceptionResNetV2, MobileNet, VGG16 and VGG19. The Deep Convolutional Autoencoder outperforms the other models by 52%, 43.55%, 13.77%, 6.67%, 22.22%, 15.11%, 6.66%, 4.89%, and 9.78% respectively. It demonstrates the efficacy of this systematic study in identifying major carps.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:0168-1699
1872-7107
DOI:10.1016/j.compag.2022.106810