One-stage CNN detector-based benthonic organisms detection with limited training dataset

Chen, Tingkai; Ning, Wang; Wang, Rongfeng; Zhao, Hong

doi:10.1016/j.neunet.2021.08.014

Cited by 42 publications

(18 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The insufficiency of training data is apparent in many fields, such as medicine and fault detection 29 , 30 . The complex structure and large number of trainable parameters of state-of-the-art classification models 31 – 34 often result in overfitting 35 . Additionally, the existing state-of-the-art classification networks cannot appropriately extract useful features from small-scale images and often exhibit poor performance on these data.…”

Section: Introductionmentioning

confidence: 99%

Consecutive multiscale feature learning-based image classification model

Olimov¹,

Subramanian

Ugli

et al. 2023

Sci Rep

View full text Add to dashboard Cite

Extracting useful features at multiple scales is a crucial task in computer vision. The emergence of deep-learning techniques and the advancements in convolutional neural networks (CNNs) have facilitated effective multiscale feature extraction that results in stable performance improvements in numerous real-life applications. However, currently available state-of-the-art methods primarily rely on a parallel multiscale feature extraction approach, and despite exhibiting competitive accuracy, the models lead to poor results in efficient computation and low generalization on small-scale images. Moreover, efficient and lightweight networks cannot appropriately learn useful features, and this causes underfitting when training with small-scale images or datasets with a limited number of samples. To address these problems, we propose a novel image classification system based on elaborate data preprocessing steps and a carefully designed CNN model architecture. Specifically, we present a consecutive multiscale feature-learning network (CMSFL-Net) that employs a consecutive feature-learning approach based on the usage of various feature maps with different receptive fields to achieve faster training/inference and higher accuracy. In the conducted experiments using six real-life image classification datasets, including small-scale, large-scale, and limited data, the CMSFL-Net exhibits an accuracy comparable with those of existing state-of-the-art efficient networks. Moreover, the proposed system outperforms them in terms of efficiency and speed and achieves the best results in accuracy-efficiency trade-off.

show abstract

Section: Introductionmentioning

confidence: 99%

Consecutive multiscale feature learning-based image classification model

Olimov¹,

Subramanian

Ugli

et al. 2023

Sci Rep

View full text Add to dashboard Cite

show abstract

“…In recent years, with the rapid development of deep learning, more and more attention has been turned to data-driven approaches. 8 Traditional data-driven approaches have certain limitations when dealing with orders of magnitude larger datasets, while machine learning can not only handle large-scale datasets but also extract hidden information from the data. And the traditional methods of manual classification and identification, traditional statistical analysis, and ocean model simulation are inefficient and imprecise for processing ocean big data.…”

Section: Introductionmentioning

confidence: 99%

Research on fault prediction of marine diesel engine based on attention-LSTM

Liu

Gan

Cong

et al. 2022

Proceedings of the Institution of Mechanical Engineers, Part M:

View full text Add to dashboard Cite

The marine diesel engine is an important guarantee for the safe operation of the ship. It is of great significance to predict the running trend of the diesel engine and infer whether the diesel engine is in an abnormal working state in advance to ensure the safe operation of the ship. In this paper, combining the feature extraction ability of the attention mechanism (Attention) and the time-series memory ability of the long short-term neural network (LSTM), a prediction model of the exhaust gas temperature of the marine diesel engine is constructed. Meanwhile, in order to improve the prediction accuracy of the model, the improved particle swarm algorithm is used to optimize the structural parameters of the model, and a model with higher prediction performance is obtained. Then, according to the residual value distribution between the predicted value and the actual value of the model, the process control method is used to set the fault threshold, and the experimental data of normal and abnormal conditions are used for verification. The results show that the proposed model can accurately predict developing failures and provide a new method for ship condition maintenance.

show abstract

“…Specifically, there are several areas that exhibit the problem of insufficiency of training data, such as medicine and fault detection 21 . Considering the complex structure and extensive number of trainable parameters of the powerful classification models [22][23][24][25] , they often result in overfitting to limited number of training data 26 . Additionally, the existing state-of-the-art classification networks cannot properly extract useful features from the small-scale images and often obtain poor performance on these kind of data.…”

Section: Introductionmentioning

confidence: 99%

CMSFL-Net: Consecutive Multi-scale Feature Learning-based Image Classification Model

Olimov

Subramanian

Ugli

et al. 2022

Preprint

View full text Add to dashboard Cite

Extracting useful features at multiple scales is one of the most crucial tasks in computer vision. Emergence of deep learning(DL) techniques and progress of convolutional neural networks (CNNs) allow powerful multi-scale feature extraction that result in stable performance increase in numerous real-life applications. However, currently available state-of-the-art powerful methods primarily rely on parallel multi-scale feature extraction approach and despite providing competitive accuracy, the models lead to poor results in efficient computation and lack at generalizing on small-scale images. On the other hand, efficient and lightweight networks cannot learn useful features properly and result in underfitting problem in case of training with small-scale images or datasets with limited number of samples. Considering these problems, we proposed a novel image classification system based on elaborate data pre-processing step and carefully designed CNN model architecture. Specifically, consecutive multiscale feature learning network (CMSFL-Net) employs consecutive feature learning approach based on usage of various feature maps with different receptive fields to obtain faster training/inference and higher accuracy. In the conducted experiments using 6 real-life image classification datasets, including small-scale, large-scale, and limited data, the proposed system obtained comparable accuracy when compared with the existing state-of-the-art powerful and efficient networks by outperforming them in terms of efficiency and speed. Moreover, the proposed system achieved the best results in accuracy-efficiency tradeoff in comparison with the currently available models.

show abstract

One-stage CNN detector-based benthonic organisms detection with limited training dataset

Cited by 42 publications

References 41 publications

Consecutive multiscale feature learning-based image classification model

Consecutive multiscale feature learning-based image classification model

Research on fault prediction of marine diesel engine based on attention-LSTM

CMSFL-Net: Consecutive Multi-scale Feature Learning-based Image Classification Model

Contact Info

Product

Resources

About