Research on target detection of Engraulis japonicus purse seine based on improved model of YOLOv5

Abstract

The refined monitoring and identification of fishing operations by fishing vessels is of great significance and value to fishing vessels. In order to solve the problem of inaccurate statistics of current Engraulis japonicus fishing quota and classification, this paper proposes an improved identification algorithm based on YOLOv5. This method introduces the SENet attention mechanism into the YOLOv5 backbone network structure, integrates the target information in different periods of fishing operations, reduces the interference of complex backgrounds, improves the precision of model detection, and ensures real-time detection efficiency. The artificially shot Engraulis japonicus videos are used as the dataset of the study, and the video is converted into a picture format to realize pre-labeling and processing. The 5550 images are divided into the training set, validation set, and test set according to 8:1:1. In order to verify the validity of the data, a set of in the control experiment, the YOLOv5 backbone network was replaced by MobileNetV2 and the SENet attention mechanism was introduced, and four models were implemented for comparison. The experimental results show that the research algorithm can obtain a mean average precision (mAP) of 99.3%, a precision of 98.9%, and a recall of 98.7%, which are improved by 1.4%, 1.7% and 2.5% respectively compared with the original model. The experimental results match the expectations. According to the statistics of some categories, the Kalman filter and the Hungarian matching method are used to count the main categories of fishing baskets, and an accuracy of 96.5% can be obtained. The threshold method for fishing nets and processing vessels can obtain 85.8% and 75% accuracy. These results shows that this target detection research can provide new ideas for job identification of Engraulis japonicus and provide auxiliary means for job statistics.

Analysis of Hauling Net Hours for Gazami Crab (Portunus trituberculatus) Gillnets Based on the Beidou Vessel Monitoring System (BDVMS)

“…The TAC value would be referred to the calculation value of the VMS. In the next step, electronic monitoring (Wang et al ., 2022; Zhang et al ., 2022) technology will be applied for the research.…”

Spatial distribution of fishing intensity of canvas stow net fishing vessels in the East China Sea and the Yellow Sea

“…The 28, 770 position data marked with the operation statuses were divided into a training set and a test set at 4: 1. The training set data was used as input into the neural network and iterated for 100 rounds to train the fishing vessels' operational status classification model (Wang et al, 2022;Zhang et al, 2022). where q(x i ) is the value, p(x i ) is the corresponding probability value.…”

Assessing the ability of deep learning techniques to perform real-time identification of shark species in live streaming video from drones

“…Advancements in computer vision and machine learning (ML) technology offer vast potential to improve upon the reliability of detections in real-time, and to automate much of the surveillance process. Modern intelligent algorithms can learn to accurately and consistently locate and identify objects in complex scenes, but do not tire like human observers and can be orders of magnitude more efficient once deployed (e.g., Longmore et al, 2017;Hodgson et al, 2018;Burr et al, 2019;Eikelboom et al, 2019;Zhang et al, 2022). Concurrently, it is expected that drone technology and associated systems (including regulations) will continue to progress rapidly, which will allow monitoring and surveillance operations to be automated end-to-end.…”

“…We believe that ML-driven shark species detectors will be excellent decision-support tools for beach managers in the very near future. The technology is already changing the way ecologists survey marine fauna (e.g., Butcher et al, 2021;Dujon et al, 2021;Jenrette et al, 2022;Marrable et al, 2022;Zhang et al, 2022;Shi et al, 2022). Humans will always need to be included in the decision loop, but the human role will change as the ML models become more reliable and flight systems become more automated.…”