Zhou Zhiguo, Sun Jiaen, Yu Jiabao, Liu Kaiyuan, Duan Junwei, Chen Long, Chen C L Philip
School of Information and Electronics, Beijing Institute of Technology, Beijing, China.
College of Information Science and Technology, Jinan University, Guangzhou, China.
Front Neurorobot. 2021 Sep 24;15:723336. doi: 10.3389/fnbot.2021.723336. eCollection 2021.
Water surface object detection is one of the most significant tasks in autonomous driving and water surface vision applications. To date, existing public large-scale datasets collected from websites do not focus on specific scenarios. As a characteristic of these datasets, the quantity of the images and instances is also still at a low level. To accelerate the development of water surface autonomous driving, this paper proposes a large-scale, high-quality annotated benchmark dataset, named Water Surface Object Detection Dataset (WSODD), to benchmark different water surface object detection algorithms. The proposed dataset consists of 7,467 water surface images in different water environments, climate conditions, and shooting times. In addition, the dataset comprises a total of 14 common object categories and 21,911 instances. Simultaneously, more specific scenarios are focused on in WSODD. In order to find a straightforward architecture to provide good performance on WSODD, a new object detector, named CRB-Net, is proposed to serve as a baseline. In experiments, CRB-Net was compared with 16 state-of-the-art object detection methods and outperformed all of them in terms of detection precision. In this paper, we further discuss the effect of the dataset diversity (e.g., instance size, lighting conditions), training set size, and dataset details (e.g., method of categorization). Cross-dataset validation shows that WSODD significantly outperforms other relevant datasets and that the adaptability of CRB-Net is excellent.
水面目标检测是自动驾驶和水面视觉应用中最重要的任务之一。迄今为止，从网站收集的现有公共大规模数据集并未专注于特定场景。作为这些数据集的一个特点，图像和实例的数量仍处于较低水平。为了加速水面自动驾驶的发展，本文提出了一个大规模、高质量标注的基准数据集，名为水面目标检测数据集（WSODD），用于对不同的水面目标检测算法进行基准测试。所提出的数据集由7467张处于不同水环境、气候条件和拍摄时间的水面图像组成。此外，该数据集总共包含14个常见目标类别和21911个实例。同时，WSODD专注于更具体的场景。为了找到一个能在WSODD上提供良好性能的简单架构，提出了一种名为CRB-Net的新目标检测器作为基线。在实验中，CRB-Net与16种先进的目标检测方法进行了比较，在检测精度方面优于所有这些方法。在本文中，我们进一步讨论了数据集多样性（如图例大小、光照条件）、训练集大小和数据集细节（如分类方法）的影响。跨数据集验证表明，WSODD明显优于其他相关数据集，并且CRB-Net的适应性非常出色。
