Multi-level residual network VGGNet for fish species classification

Prasetyo, Eko; Suciati, Nanik; Fatichah, Chastine

doi:10.1016/j.jksuci.2021.05.015

Cited by 44 publications

(21 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…So, there was no need to use image prefiltering. The results show that the use of multi-level residual network [25], machine learning algorithm [17], SURF [11] achieved high accuracy with 99.69%, 98% and 98.67% respectively. Maximum number of classified fish species was 38 species [17].…”

Section: Discussionmentioning

confidence: 94%

“…Early techniques for fish types identification were carried out in controlled circumference only. Most of the researchers attempt to identify the fish images based on: the off-line fish images, natural environmental conditions and databases in [15,21,23,25,28]. So, there was no need to use image prefiltering.…”

Section: Discussionmentioning

confidence: 99%

“…[21] Genetic algorithm with tabu search and a back-propagation algorithm 82.1% for back-propagation, 87% for GTB [23] Ma-b algorithm and back propagation algorithm 90% for MA-B algorithm 82.25% for back propagation algorithm [24] Deep convolutional neural networks 80.58% [25] Multi-level residual network 99.69% [28] Neural network Not reported [32] Convolutional neural networks 96.55%.…”

Section: Reference Techniques Accuracymentioning

confidence: 99%

See 2 more Smart Citations

Fish Species Identification Techniques: A Review

Hassoon¹

2022

ANJS

View full text Add to dashboard Cite

Fish species identification process is still problematical and time consuming, so automated fish identification is important. The performance of various fish identification techniques is compared depending on pre-processing, number of significant characteristics and identification correctness. The reason for writing this review stems from the importance of fish for humans as food and as a wealth which support the economy of countries. Therefore, it is necessary to shed light on fish and how to distinguish the beneficial and harmful ones, discover and identify fish diseases, and so on, which contributes in building a knowledge base available to all. The aim of this paper is present a review on various fish species identification techniques; pre-processing operations, features extraction methods, recognition/classification techniques and databases. This review will be helpful for beginner researchers, it can be used as a starting point for new fish species identification approaches.

show abstract

Section: Discussionmentioning

confidence: 94%

Section: Discussionmentioning

confidence: 99%

Section: Reference Techniques Accuracymentioning

confidence: 99%

See 1 more Smart Citation

Fish Species Identification Techniques: A Review

Hassoon¹

2022

ANJS

View full text Add to dashboard Cite

show abstract

“…The third step is enforced image normalization, dividing the intensity value of each pixel by 255. The reason is to reduce computational costs because the intensity value of each pixel is between 0 and 1 [41].…”

Section: Datasetmentioning

confidence: 99%

Convolutional Neural Network (CNN) Models for Crop Diseases Classification

Sutaji

Rosyid

2022

KINETIK

View full text Add to dashboard Cite

Crop diseases have a significant impact on agricultural production. As a result, early diagnosis of crop diseases is critical. Deep learning approaches are now promising to improve disease detection. Convolutional Neural Network (CNN) models can detect crop disease using images with automatic feature extraction. This study proposes crop disease classification considering ten pre-trained CNN models. Fine-tuning for each model was conducted in the Plant Village dataset. The experimental results show that fine-tuning improves the model’s performance with an average accuracy of 8.85%. The best CNN model was DenseNet121, with 94.48% and 98.97% accuracy for freezing all layers and unfreezing last block convolution layers. Moreover, fine-tuning produces less time-consuming with an average of 2.20 hours. VGG19 is the less time-consuming reduction by 8 hours. On the other hand, MobileNetV2 is the second-best performance model with less time-consuming than DenseNet121, and produces fewer parameters, which is affordable for embedding it to mobile devices.

show abstract

“…(2018) analysed the performance of three networks: AlexNet, GoogleNet, and LeNet. Prasetyo et al, (2021) proposed a new residual network strategy called MLR (Multi-level residual) by combining low-level features with high-level features using depth-wise separable convolution (DSC) for the FishKnowledge dataset. Zhang et al (2021) extracted texture features after reducing the noise from the fish4knowledge dataset and implemented a Deep Neural Network (DNN).…”

Section: Introductionmentioning

confidence: 99%

Multi-species Fish Identification using Hybrid DeepCNN with Refined Squeeze and Excitation Architecture

Veluswami¹,

Panneerselvam²

2022

ase

View full text Add to dashboard Cite

Fish play a prominent role in the food web and fish farming has value for both human consumption and tourist attractions. Due to the increasing importance of marine biodiversity, recognition of fish species has become a prominent task in monitoring the mislabelling of seafood and extinct species. This problem can be solved using traditional manual annotation on the images. To reduce manpower, cost, and tremendous time, deep learning approaches are used which always require large datasets. Therefore, fish species identification is a challenging task using disproportionately small data sets. In this research, we develop a new method by refining the squeeze and excitation network for the automatic fish species classification model to identify 23 different types of fish species. To achieve this, a hybrid framework using deep learning is proposed on a large-scale dataset and implemented transfer learning for a small-scale dataset. Deep learning methods can be used to identify fish in underwater images. In this study, we have proposed a new method of hybrid Deep Convolutional Neural Network (CNN) along with a Support Vector Machine (SVM) for classification. Additionally, the Squeeze and Excitation (SE) block has been improved for improved feature extraction. The proposed method achieved an accuracy of 97.90%. Then post-training with the small-scale dataset (Croatian) achieved an accuracy of 94.99% with an 11% improvement compared to Bilinear CNN (B-CNN) (Qui et al., 2018) and can be used in any underwater applications to identify fish species and avoid mislabelling of seafood.

show abstract

Multi-level residual network VGGNet for fish species classification

Cited by 44 publications

References 22 publications

Fish Species Identification Techniques: A Review

Fish Species Identification Techniques: A Review

Convolutional Neural Network (CNN) Models for Crop Diseases Classification

Multi-species Fish Identification using Hybrid DeepCNN with Refined Squeeze and Excitation Architecture

Contact Info

Product

Resources

About