Lightweight Underwater Object Detection Based on YOLO v4 and Multi-Scale Attentional Feature Fusion

A challenging and attractive task in computer vision is underwater object detection. Although object detection techniques have achieved good performance in general datasets, problems of low visibility and color bias in the complex underwater environment have led to generally poor image quality; besi...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Minghua Zhang, Shubo Xu, Wei Song, Qi He, Quanmiao Wei
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	YOLO lightweight network underwater object detection attention mechanism Science Q
Acceso en línea:	https://doaj.org/article/32892cbc140549528f55d2098c781f1e
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:32892cbc140549528f55d2098c781f1e
record_format	dspace
spelling	oai:doaj.org-article:32892cbc140549528f55d2098c781f1e2021-11-25T18:55:37ZLightweight Underwater Object Detection Based on YOLO v4 and Multi-Scale Attentional Feature Fusion10.3390/rs132247062072-4292https://doaj.org/article/32892cbc140549528f55d2098c781f1e2021-11-01T00:00:00Zhttps://www.mdpi.com/2072-4292/13/22/4706https://doaj.org/toc/2072-4292A challenging and attractive task in computer vision is underwater object detection. Although object detection techniques have achieved good performance in general datasets, problems of low visibility and color bias in the complex underwater environment have led to generally poor image quality; besides this, problems with small targets and target aggregation have led to less extractable information, which makes it difficult to achieve satisfactory results. In past research of underwater object detection based on deep learning, most studies have mainly focused on improving detection accuracy by using large networks; the problem of marine underwater lightweight object detection has rarely gotten attention, which has resulted in a large model size and slow detection speed; as such the application of object detection technologies under marine environments needs better real-time and lightweight performance. In view of this, a lightweight underwater object detection method based on the MobileNet v2, You Only Look Once (YOLO) v4 algorithm and attentional feature fusion has been proposed to address this problem, to produce a harmonious balance between accuracy and speediness for target detection in marine environments. In our work, a combination of MobileNet v2 and depth-wise separable convolution is proposed to reduce the number of model parameters and the size of the model. The Modified Attentional Feature Fusion (AFFM) module aims to better fuse semantic and scale-inconsistent features and to improve accuracy. Experiments indicate that the proposed method obtained a mean average precision (mAP) of 81.67% and 92.65% on the PASCAL VOC dataset and the brackish dataset, respectively, and reached a processing speed of 44.22 frame per second (FPS) on the brackish dataset. Moreover, the number of model parameters and the model size were compressed to 16.76% and 19.53% of YOLO v4, respectively, which achieved a good tradeoff between time and accuracy for underwater object detection.Minghua ZhangShubo XuWei SongQi HeQuanmiao WeiMDPI AGarticleYOLOlightweight networkunderwater object detectionattention mechanismScienceQENRemote Sensing, Vol 13, Iss 4706, p 4706 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	YOLO lightweight network underwater object detection attention mechanism Science Q
spellingShingle	YOLO lightweight network underwater object detection attention mechanism Science Q Minghua Zhang Shubo Xu Wei Song Qi He Quanmiao Wei Lightweight Underwater Object Detection Based on YOLO v4 and Multi-Scale Attentional Feature Fusion
description	A challenging and attractive task in computer vision is underwater object detection. Although object detection techniques have achieved good performance in general datasets, problems of low visibility and color bias in the complex underwater environment have led to generally poor image quality; besides this, problems with small targets and target aggregation have led to less extractable information, which makes it difficult to achieve satisfactory results. In past research of underwater object detection based on deep learning, most studies have mainly focused on improving detection accuracy by using large networks; the problem of marine underwater lightweight object detection has rarely gotten attention, which has resulted in a large model size and slow detection speed; as such the application of object detection technologies under marine environments needs better real-time and lightweight performance. In view of this, a lightweight underwater object detection method based on the MobileNet v2, You Only Look Once (YOLO) v4 algorithm and attentional feature fusion has been proposed to address this problem, to produce a harmonious balance between accuracy and speediness for target detection in marine environments. In our work, a combination of MobileNet v2 and depth-wise separable convolution is proposed to reduce the number of model parameters and the size of the model. The Modified Attentional Feature Fusion (AFFM) module aims to better fuse semantic and scale-inconsistent features and to improve accuracy. Experiments indicate that the proposed method obtained a mean average precision (mAP) of 81.67% and 92.65% on the PASCAL VOC dataset and the brackish dataset, respectively, and reached a processing speed of 44.22 frame per second (FPS) on the brackish dataset. Moreover, the number of model parameters and the model size were compressed to 16.76% and 19.53% of YOLO v4, respectively, which achieved a good tradeoff between time and accuracy for underwater object detection.
format	article
author	Minghua Zhang Shubo Xu Wei Song Qi He Quanmiao Wei
author_facet	Minghua Zhang Shubo Xu Wei Song Qi He Quanmiao Wei
author_sort	Minghua Zhang
title	Lightweight Underwater Object Detection Based on YOLO v4 and Multi-Scale Attentional Feature Fusion
title_short	Lightweight Underwater Object Detection Based on YOLO v4 and Multi-Scale Attentional Feature Fusion
title_full	Lightweight Underwater Object Detection Based on YOLO v4 and Multi-Scale Attentional Feature Fusion
title_fullStr	Lightweight Underwater Object Detection Based on YOLO v4 and Multi-Scale Attentional Feature Fusion
title_full_unstemmed	Lightweight Underwater Object Detection Based on YOLO v4 and Multi-Scale Attentional Feature Fusion
title_sort	lightweight underwater object detection based on yolo v4 and multi-scale attentional feature fusion
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/32892cbc140549528f55d2098c781f1e
work_keys_str_mv	AT minghuazhang lightweightunderwaterobjectdetectionbasedonyolov4andmultiscaleattentionalfeaturefusion AT shuboxu lightweightunderwaterobjectdetectionbasedonyolov4andmultiscaleattentionalfeaturefusion AT weisong lightweightunderwaterobjectdetectionbasedonyolov4andmultiscaleattentionalfeaturefusion AT qihe lightweightunderwaterobjectdetectionbasedonyolov4andmultiscaleattentionalfeaturefusion AT quanmiaowei lightweightunderwaterobjectdetectionbasedonyolov4andmultiscaleattentionalfeaturefusion
_version_	1718410514451660800

Lightweight Underwater Object Detection Based on YOLO v4 and Multi-Scale Attentional Feature Fusion

Ejemplares similares