SWIPENET: Object detection in noisy underwater images
This addresses the problem of robust object detection for underwater robotics applications, offering an incremental improvement by adapting existing deep learning methods to handle domain-specific challenges like noise and small objects.
The paper tackles object detection in noisy underwater images, where blur and noise degrade performance, by proposing SWIPENET with high-resolution feature maps for small objects and a sample-weighted loss, combined with a Curriculum Multi-Class Adaboost training paradigm to handle noise, achieving better accuracy on URPC2017 and URPC2018 datasets compared to state-of-the-art methods.
In recent years, deep learning based object detection methods have achieved promising performance in controlled environments. However, these methods lack sufficient capabilities to handle underwater object detection due to these challenges: (1) images in the underwater datasets and real applications are blurry whilst accompanying severe noise that confuses the detectors and (2) objects in real applications are usually small. In this paper, we propose a novel Sample-WeIghted hyPEr Network (SWIPENET), and a robust training paradigm named Curriculum Multi-Class Adaboost (CMA), to address these two problems at the same time. Firstly, the backbone of SWIPENET produces multiple high resolution and semantic-rich Hyper Feature Maps, which significantly improve small object detection. Secondly, a novel sample-weighted detection loss function is designed for SWIPENET, which focuses on learning high weight samples and ignore learning low weight samples. Moreover, inspired by the human education process that drives the learning from easy to hard concepts, we here propose the CMA training paradigm that first trains a clean detector which is free from the influence of noisy data. Then, based on the clean detector, multiple detectors focusing on learning diverse noisy data are trained and incorporated into a unified deep ensemble of strong noise immunity. Experiments on two underwater robot picking contest datasets (URPC2017 and URPC2018) show that the proposed SWIPENET+CMA framework achieves better accuracy in object detection against several state-of-the-art approaches.