Underwater target recognition using convolutional recurrent neural networks with 3-D Mel-spectrogram and data augmentation

Liu, Feng; Shen, Tongsheng; Luo, Zailei; Zhao, Dexin; Guo, Sen

doi:10.1016/j.apacoust.2021.107989

Cited by 91 publications

(39 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The MFCC performs discrete cosine transform (DCT) on the Mel spectrogram [43]. This process reduces abnormal values in the respiratory sounds and removes noise by compressing it [44]. Therefore, in this study, the MFCC was applied, and the MFCC attribute extraction estimates had several stages, which are explained below:…”

Section: ) Log-mel Spectrogram-mfccmentioning

confidence: 99%

Lightweight Skip Connections With Efficient Feature Stacking for Respiratory Sound Classification

Choi

Lee

et al. 2022

IEEE Access

View full text Add to dashboard Cite

As the number of deaths from respiratory diseases due to COVID-19 and infectious diseases increases, early diagnosis is necessary. In general, the diagnosis of diseases is based on imaging devices (e.g., computed tomography and magnetic resonance imaging) as well as the patient's underlying disease information. However, these examinations are time-consuming, incur considerable costs, and in a situation like the ongoing pandemic, face-to-face examinations are difficult to conduct. Therefore, we propose a lung disease classification model based on deep learning using non-contact auscultation. In this study, two respiratory specialists collected normal respiratory sounds and five types of abnormal sounds associated with lung disease, including those associated with four lung lesions in the left and right anterior chest and left and right posterior chest. For preprocessing and feature extraction, the noise was removed using three pass filters (low, band, and high), and respiratory sound features were extracted using the Log-Mel Spectrogram-Mel Frequency Cepstral Coefficient followed by feature stacking. Then, we propose a lung disease classification model of dense lightweight convolutional neural network-bidirectional gated recurrent unit skip connections using depthwise separable convolution based on the extracted respiratory sound information. The performance of the classification model was compared with both the baseline and the lightweight models. The results indicate that the proposed model achieves high performance and has an accuracy of 92.3%, sensitivity of 92.1%, specificity of 98.5%, and f1-score of 91.9%. Using the proposed model, we aim to contribute to the early detection of diseases during the COVID-19 pandemic.

show abstract

Section: ) Log-mel Spectrogram-mfccmentioning

confidence: 99%

Lightweight Skip Connections With Efficient Feature Stacking for Respiratory Sound Classification

Choi

Lee

et al. 2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…In Figure 9 , the used network is compared with DNN [ 13 ], ADCNN [ 15 ], DepthCNN [ 16 ], CDBN [ 17 ], and CRNN [ 18 ]. HRNet is the hybrid routing network method.…”

Section: Methodsmentioning

confidence: 99%

“…The deep-belief network utilizes the pre-processing approach of standard Boltzmann machine, the hidden middle layer of the clustering method, and the training optimization of parameter updating [ 17 ]. The underwater acoustic signal classifier adopts the convolutional recurrent neural network (CRNN), and the recurrent neural network (RNN), combined with a CNN, to acquire the different natures of sound characteristics, which further enhances the recognition effects by data augmentation [ 18 ].…”

Section: Introductionmentioning

confidence: 99%

Underwater Target Signal Classification Using the Hybrid Routing Neural Network

Cheng

Zhang

2021

Sensors

View full text Add to dashboard Cite

In signal analysis and processing, underwater target recognition (UTR) is one of the most important technologies. Simply and quickly identify target types using conventional methods in underwater acoustic conditions is quite a challenging task. The problem can be conveniently handled by a deep learning network (DLN), which yields better classification results than conventional methods. In this paper, a novel deep learning method with a hybrid routing network is considered, which can abstract the features of time-domain signals. The used network comprises multiple routing structures and several options for the auxiliary branch, which promotes impressive effects as a result of exchanging the learned features of different branches. The experiment shows that the used network possesses more advantages in the underwater signal classification task.

show abstract

“…The underwater image recognition process in the traditional manual feature period is divided into regional selection, feature extraction and classifier classification, as shown in Figure 1. Underwater image recognition is a practical application of target recognition [8], which also includes feature extraction of images. According to the actual situation under water, features such as color, texture, shape, etc.…”

Section: Period Of Traditional Manual Featuresmentioning

confidence: 99%

A Literature Review of Underwater Image Detection

Wang

Jin

et al. 2022

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

As one of the world’s few large-scale aquaculture countries, the mechanization, output and efficiency are low in Chinese aquaculture industry.with the development of industrial automation technology and computer image processing technology, aquaculture will be automated, scientific management transformation. As a specific application of computer recognition technology, underwater image recognition technology is important to promote aquaculture industry automation and intelligence. In this paper, first of all introduces the existing research of underwater target image recognition, mainly presents the underwater image recognition technology based on deep learning. And then the current problems of underwater image recognition is summarized. Besides, the future development of underwater image recognition technology and its application in wisdom fishery is also discussed, which provides other scholar with some theoretical help and a comprehensive reference.

show abstract

Underwater target recognition using convolutional recurrent neural networks with 3-D Mel-spectrogram and data augmentation

Cited by 91 publications

References 24 publications

Lightweight Skip Connections With Efficient Feature Stacking for Respiratory Sound Classification

Lightweight Skip Connections With Efficient Feature Stacking for Respiratory Sound Classification

Underwater Target Signal Classification Using the Hybrid Routing Neural Network

A Literature Review of Underwater Image Detection

Contact Info

Product

Resources

About