A Novel Underwater Image Enhancement and Improved Underwater Biological Detection Pipeline

  title={A Novel Underwater Image Enhancement and Improved Underwater Biological Detection Pipeline},
  author={Zheng Liu and Yaoming Zhuang and Pengrun Jia and Chengdong Wu and Hongli Xu and Zhanlin Liu},
For aquaculture resource evaluation and ecological environment monitoring, automatic detection and identification of marine organisms is critical. However, due to the low quality of underwater images and the characteristics of underwater biological, a lack of abundant features may impede traditional hand-designed feature extraction approaches or CNN-based object detection algorithms, particularly in complex underwater environment. Therefore, the goal of this paper is to perform object detection… 


Composited FishNet: Fish Detection and Species Recognition From Low-Quality Underwater Videos
A novel composite fish detection framework based on a composite backbone and an enhanced path aggregation network called Composited FishNet is proposed, which enhances the characteristic information output of the detected object and improves the utilization of characteristic information.
Dual Refinement Underwater Object Detection Network
The developed underwater object detection framework has a composite connection backbone to boost the feature representation and introduces a receptive field augmentation module to exploit multi-scale contextual features and provides a prediction refinement scheme according to six prediction layers.
Lightweight Deep Neural Network for Joint Learning of Underwater Object Detection and Color Conversion.
The presented experimental results with the implementation on the Raspberry pi platform have justified the effectiveness of the proposed lightweight jointly learning model for underwater object detection compared with the state-of-the-art approaches.
Underwater image enhancement via extended multi-scale Retinex
Real-World Underwater Enhancement: Challenges, Benchmarks, and Solutions Under Natural Light
The object detection performance on enhanced images is exploited as a brand new task-specific evaluation criterion and suggested promising solutions and new directions for visibility enhancement, color correction, and object detection on real-world underwater images are suggested.
Perceptual Underwater Image Enhancement With Deep Learning and Physical Priors
A hybrid underwater image synthesis model, which fuses physical priors and data-driven cues, is proposed to synthesise training data and generalise the authors' enhancement model for real-world underwater images.
Automatic Red-Channel underwater image restoration
Color Balance and Fusion for Underwater Image Enhancement
This work introduces an effective technique to enhance the images captured underwater and degraded due to the medium scattering and absorption by building on the blending of two images that are directly derived from a color-compensated and white-balanced version of the original degraded image.
Enhancing underwater images and videos by fusion
A novel strategy to enhance underwater videos and images that derives the inputs and the weight measures only from the degraded version of the image, and supports temporal coherence between adjacent frames by performing an effective edge preserving noise reduction strategy.
Fast accurate fish detection and recognition of underwater images with Fast R-CNN
Fast R-CNN improves mean average precision (mAP) by 11.2% relative to Deformable Parts Model (DPM) baseline-achieving a mAP of 81.4%, and detects 80× faster than previous R- CNN on a single fish image.