Long-yu Jiang

Prof. Long-yu Jiang

orcid

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
2 June 2020 in IEEE Access
Verified
Journal Article
Open access
LJLong-yu Jiang
Long-yu Jiang
TCTao Cai
Tao Cai
QMQixiang Ma
Qixiang Ma
FXFanjin Xu
Fanjin Xu
SWShijie Wang
Shijie Wang
Object detection in sonar images has always been a challenge due to the low resolution of sonar images and the strong noise existing in them. Although convolutional neural networks (CNNs) have been applied to detect objects in sonar images, successful detection is impeded by the lack of large annotated sonar images. Manual annotation is not only tedious and time consuming but also demands specialty-oriented knowledge and skills, which are not easily accessible. To dramatically reduce the heavy annotation cost, this paper proposes three simple but effective active-learning-based algorithms for object detection, which can reduce the annotation cost by seeking the most informative images from unlabeled data and then continuously retraining a model by merging newly annotated samples in each iteration into an already labeled dataset to enhance the CNN’s performance. The results of the experiments illustrate that the proposed active framework with approximately 35% data can achieve competitive results compared to the CNN’s performance using all data.

30 September 2024 in Robotics and Autonomous Systems
Verified
Journal Article
Closed access
CYChao Yang
Chao Yang
LJLong-yu Jiang
Long-yu Jiang
ZLZhicheng Li
Zhicheng Li
JWJie Wu
Jie Wu

29 June 2022 in Electronics
Verified
Journal Article
Open access
PDF file
FXFangjin Xu
Fangjin Xu
JHJianxing Huang
Jianxing Huang
JWJie Wu
Jie Wu
LJLong-yu Jiang
Long-yu Jiang
Instance segmentation of sonar images is an effective method for underwater target recognition. However, the mismatch among positioning accuracy found by boxIoU and classification confidence, which is used as NMS score in current instance segmentation models; and the high annotation cost of sonar images, are two major problems in the task. To tackle these problems, in this paper, we present a novel instance segmentation method called Mask-Box Scoring R-CNN and embedded it in our proposed deep active learning framework. For the mismatch problem between boxIoU and NMS score, Mask-Box Scoring R-CNN uses a boxIoU head to predict the quality of the bounding boxes. We amend the non-maximum suppression (NMS) score predicted by BoxIoU to preserve high-quality bounding boxes in inference flow. To deal with the annotating problem, we propose a triplets-measure-based active learning (TBAL) method and a balanced-sampling method applicable for deep learning. The TBAL method evaluates the amount of information of unlabeled samples from the aspects of classification confidence, positioning accuracy, and mask quality. The balanced-sampling method selects hard samples from the dataset to train the model to improve performance. The experimental results show that Mask-Box Scoring R-CNN achieves improvements of 1% in boxAP and 1.3% boxAP on our sonar image dataset compared with Mask Scoring R-CNN and Mask R-CNN, respectively. The active learning framework with TBAL and balanced sampling can achieve a competitive performance with less labeled samples than other frameworks, which can better facilitate underwater target recognition.

1 June 2023 in Machine Vision and Applications
Verified
Journal Article
Closed access
PDF file
CYChao Yang
Chao Yang
YLYongpeng Li
Yongpeng Li
LJLong-yu Jiang
Long-yu Jiang
JHJianxing Huang
Jianxing Huang

24 February 2017 in IEEE Journal of Oceanic Engineering
Verified
Journal Article
Closed access
LJLong-yu Jiang
Long-yu Jiang
Philippe RouxPhilippe Roux
Philippe Roux
JMJerome I. Mars
Jerome I. Mars
One intriguing property that is exploited by ocean acoustic tomography (OAT) is that acoustic signals travel in a multipath. As the first step of OAT, each raypath should be identified with a particular travel time. However, the set of multipath rays generated by an emitted signal is correlated or coherent, as they are produced by reflection and or by refraction in propagation. In this paper, a high-resolution method called smoothing multiple signal classification active large band (MUSICAL) is presented in the context of shallow-water OAT for separating coherent or fully correlated raypaths in the direction-of-arrival temporal domain. The method is a combination of the MUSICAL and spatial-frequency smoothing processing. Furthermore, the performance of smoothing-MUSICAL is illustrated by experiments based on both synthetic data and real data. This algorithm largely improves separation performances and presents fewer artifacts compared with conventional beamforming. In particular, experimental results show that smoothing-MUSICAL is more robust than beamforming facing a noisy environment with moderate signal-to-noise ratio.

1 April 2018 in The Journal of the Acoustical Society of America
Verified
Journal Article
Open access
PDF file
LJLong-yu Jiang
Long-yu Jiang
WSWenbo Song
Wenbo Song
ZZZhe Zhang
Zhe Zhang
CYChunfeng Yang
Chunfeng Yang
SWShijie Wang
Shijie Wang
PRPhilippe Roux
Philippe Roux
Subspace algorithms based on higher-order cumulants were developed to achieve high-resolution separation in non-Gaussian processes. However, singular value decomposition (SVD) of a huge matrix is an unavoidable step of these algorithms. The memory space and running time required by the decomposition are super-linear with respect to the size of the matrix, which is prohibitive in terms of practical applications. Thus, in this paper, a fast raypath separation algorithm based on low-rank matrix approximation is proposed in a shallow-water waveguide. The experimental results illustrate that the proposed algorithm dramatically reduces the consumption of time and space, with arbitrarily small error, compared to conventional higher-order cumulant-based algorithms.