Wei Zhao

Prof. Wei Zhao

Beihang University
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
1 October 2015 in IET Image Processing
Verified
Journal Article
Open access
PDF file
HLHuang Lidong
Huang Lidong
WZWei Zhao
Wei Zhao
WJWang Jun
Wang Jun
SZSun Zebin
Sun Zebin
Image enhancement has an important role in image processing applications. Contrast limited adaptive histogram equalisation (CLAHE) is an effective algorithm to enhance the local details of an image. However, it faces the contrast overstretching and noise enhancement problems. To solve these problems, this study presents a novel image enhancement method, named CLAHE‐discrete wavelet transform (DWT), which combines the CLAHE with DWT. The new method includes three main steps: First, the original image is decomposed into low‐frequency and high‐frequency components by DWT. Then, the authors enhance the low‐frequency coefficients using CLAHE and keep the high‐frequency coefficients unchanged to limit noise enhancement. This is because the high‐frequency component corresponds to the detail information and contains most noises of original image. Finally, reconstruct the image by taking inverse DWT of the new coefficients. To alleviate over‐enhancement, the reconstructed and original images are averaged using an originally proposed weighting factor. The weighting operation can control the enhancement levels of regions with different luminances in original image adaptively. This is important because bright parts of image are usually needless to be enhanced in comparison with the dark parts. Extensive experiments show that this method performs well in detail preservation and noise suppression.

24 February 2022 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
Xudong FanXudong Fan
Xudong Fan
WZWei Zhao
Wei Zhao
Automatic license plate recognition plays an important role in intelligent transportation systems and is of great significance. However, at present, most current approaches are only concerned of license plate recognition under restrictive conditions, where the license plates are shot in a frontal view and under good light conditions. These approaches are not robust enough in real-world complex capture scenarios, such as uneven light condition or oblique shooting angle. In order to improve the robustness of recognizing license plates under complex capture scenarios, a robust license plate detection network (CA-CenterNet) is proposed in this paper, together with a segmentation-free network (CNNG) for the recognition of license plate characters. CA-CenterNet can detect not only the center of each license plate, but also four vectors pointing to the four corners of the corresponding license plate, regardless of the rotation and distortion of the license plates, which gives us the possibility to rectify the distorted license plates in the source images. Then, CNNG can accurately identify the characters in the detected license plates without character segmentation. Experimental results prove that our automatic license plate recognition system has good performance in real-world complex capture scenarios and outperforms current license plate recognition models.

17 January 2025 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
XSXiping Shang
Xiping Shang
NLNannan Li
Nannan Li
DLDongjin Li
Dongjin Li
JLJianwei Lv
Jianwei Lv
WZWei Zhao
Wei Zhao
JXJingyu Xu
Jingyu Xu
Vehicle detection based on remote sensing images is widely used in urban traffic management and disaster rescue. RGB images, which are used more, lead to poor detection performance in low light conditions due to the imaging mechanism. At present, the main solution is to improve the detection performance in low light by fusing with infrared images. However, the current methods often overlook the impact of illumination changes on RGB images, and ignore the important role of high-frequency information for object detection, especially for low-light target detection. In this paper, we propose a Cross-modality and Cross-domain Low-light Detector (CCLDet) for low-light vehicle detection, including three improvements. First, an object illumination-aware module (OIAM) is proposed, which can adjust adaptively the weight of different modalities according to the object illumination intensity in the training phase and enables the detector to adapt to different lighting conditions. Second, we propose a visibility loss, which converts the position deviation into the illumination intensity deviation of each point in the object area. Compared with relying only on semantic information for object localization, the illumination makes the information that can be used for localization more abundant. Third, we design a cross-domain feature fusion module (CDFFM), which can enhance high-frequency features and enrich target information when low-frequency features are lost due to low light pollution. Extensive experiments on three challenging RGB-infrared objects detection datasets demonstrated the mAP and the parameter quantities of CCLDet over popular object detectors.

1 March 2015 in IET Image Processing
Verified
Journal Article
Open access
PDF file
WZWei Zhao
Wei Zhao
HLHuang Lidong
Huang Lidong
WJWang Jun
Wang Jun
SZSun Zebin
Sun Zebin
Contrast enhancement plays an important role in image processing applications. The global histogram equalisation (GHE)‐based techniques are very popular for their simpleness. In the author's study, the authors originally divide the GHE techniques into two steps, that is, the pixel populations mergence (PPM) step and the grey‐levels distribution (GLD) step. In the PPM step, the pixel populations of adjoining grey scales to be mapped to the same grey scale are merged firstly in input histogram. Then, the new grey scales are redistributed according to a corresponding transformation function in the GLD step. This division is meaningful because the entropy of enhanced image is only determined by pixel populations regardless of grey levels. Then, they prove the entropy of enhanced image is reduced because of mergence. Inspired by GHE, they propose a novel entropy maximisation histogram modification scheme, which also consists of PPM and GLD steps. However, the entropy is maximised, that is, the reduction of entropy is minimised under originally presented entropy maximisation rule in their PPM step. In the GLD step, they redistribute the grey scales in the merged histogram using a log‐based distribution function to control the enhancement level. Experimental results demonstrate the proposed method is effective.

17 June 2020 in IEEE Geoscience and Remote Sensing Letters
Verified
Journal Article
Closed access
YWYang Wang
Yang Wang
ZSZhaochen Sun
Zhaochen Sun
WZWei Zhao
Wei Zhao
With the development of convolutional neural networks, the semantic segmentation of remote sensing images has been widely developed, but there are still some unsolved problems in this field due to the lack of multiscale information and the feature mismatch at the upsampling process. To solve these problems, we propose a network called multiscale feature fusion and alignment network (MFANet). MFANet is composed of an encoder and a decoder. The encoder contains a fully convolutional network, a multilevel feature fusion block (MLFFB), and a multiscale feature pyramid (MSFP). These subnetworks can obtain fine-grained feature maps that are full of multiscale and global features and improve segmentation results at multiple object scales. Moreover, MFANet uses a light convolution subnetwork, called decoder, to upsample the segmentation map stage by stage. Combining three scales of features, the decoder can promote the feature alignment at the upsampling stage. Along with the decoder, MFANet utilizes a multistage supervision loss to enhance the localization performance and boundary regression ability. Benefitting from the encoder and decoder structure and the innovative components inside encoder, MFANet is very powerful for the semantic segmentation of remote sensing images and can suit the complicated environment. We evaluate our MFANet on the Vaihingen and Potsdam data sets, and it outperforms the state-of-art methods both in the metric and visual effect.

15 November 2020 in Sensors
Verified
Journal Article
Open access
PDF file
Ruihong YinRuihong Yin
Ruihong Yin
WZWei Zhao
Wei Zhao
Xudong FanXudong Fan
Xudong Fan
Yongfeng YinYongfeng Yin
Yongfeng Yin
There are a large number of studies on geospatial object detection. However, many existing methods only focus on either accuracy or speed. Methods with both fast speed and high accuracy are of great importance in some scenes, like search and rescue, and military information acquisition. In remote sensing images, there are some targets that are small and have few textures and low contrast compared with the background, which impose challenges on object detection. In this paper, we propose an accurate and fast single shot detector (AF-SSD) for high spatial remote sensing imagery to solve these problems. Firstly, we design a lightweight backbone to reduce the number of trainable parameters of the network. In this lightweight backbone, we also use some wide and deep convolutional blocks to extract more semantic information and keep the high detection precision. Secondly, a novel encoding–decoding module is employed to detect small targets accurately. With up-sampling and summation operations, the encoding–decoding module can add strong high-level semantic information to low-level features. Thirdly, we design a cascade structure with spatial and channel attention modules for targets with low contrast (named low-contrast targets) and few textures (named few-texture targets). The spatial attention module can extract long-range features for few-texture targets. By weighting each channel of a feature map, the channel attention module can guide the network to concentrate on easily identifiable features for low-contrast and few-texture targets. The experimental results on the NWPU VHR-10 dataset show that our proposed AF-SSD achieves superior detection performance: parameters 5.7 M, mAP 88.7%, and 0.035 s per image on average on an NVIDIA GTX-1080Ti GPU.