Kunlong ZHENG

Mr. Kunlong ZHENG

College of Information Engineering, Inner Mongolia University of Technology, Hohhot, Inner Mongolia, China
orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
1 December 2022 in Remote Sensing
Verified
Journal Article
Open access
PDF file
KZKunlong ZHENG
Kunlong ZHENG
YDYifan Dong
Yifan Dong
WXWei Xu
Wei Xu
SYSu Yun
Su Yun
PHPingping Huang
Pingping Huang
In recent years, dramatic progress in object detection in remote sensing images has been made due to the rapid development of convolutional neural networks (CNNs). However, most existing methods solely pay attention to training a suitable network model to extract more powerful features in order to solve the problem of false detections and missed detections caused by background complexity, various scales, and the appearance of the object. To open up new paths, we consider embedding knowledge into geospatial object detection. As a result, we put forward a method of digitizing knowledge and embedding knowledge into detection. Specifically, we first analyze the training set and then transform the probability into a knowledge factor according to an analysis using an improved version of the method used in existing work. With a knowledge matrix consisting of knowledge factors, the Knowledge Inference Module (KIM) optimizes the classification in which the residual structure is introduced to avoid performance degradation. Extensive experiments are conducted on two public remote sensing image data sets, namely DOTA and DIOR. The experimental results prove that the proposed method is able to reduce some false detections and missed detections and obtains a higher mean average precision (mAP) performance than the baseline method.

15 August 2023 in IEEE Geoscience and Remote Sensing Letters
Verified
Journal Article
Closed access
KZKunlong ZHENG
Kunlong ZHENG
YDYifan Dong
Yifan Dong
WXWei Xu
Wei Xu
WTWeixian Tan
Weixian Tan
PHPingping Huang
Pingping Huang
Object detection in remote sensing images (RSIs) is crucial for ground observation applications such as land surveying, urban planning, and precision agriculture. With the advent of convolutional neural networks, the performance of object detection in RSIs has been significantly improved. However, most studies have focused exclusively on the feature extraction, processing, and representation of geographical objects. The lack of integration of relevant common-sense knowledge has led to some obvious missed detections and illogical false alarms in the detection results. This letter proposes a novel and widely applicable approach for updating and fusing object co-occurrence knowledge into a two-stage multi-class object detection framework, called Auto Learner of Co-occurrence. The method captures the correlation between different object categories during the training process with the Co-occurrence Knowledge Updating module and utilizes the established co-occurrence weight matrix to enhance classification with the Knowledge Fusing module. The effectiveness and adaptability of the proposed method were validated through experiments on two multi-class object detection datasets using four popular two-stage detectors. The experiments also indicate that the speed penalty and the parameter increase resulting from the method are negligible.