WHWei Huang

Wei Huang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Wei Huang has a total of 17 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Spatial–Adaptive Replay for Foreground Classes in Class-Incremental Semantic Segmentation

    by Wei Huang, Zhuoming Gu, Mengfan Xu, Xiaofeng Lu

    Class-Incremental Semantic Segmentation (CISS) addresses the challenge of catastrophic forgetting in semantic segmentation models. In autonomous driving scenarios, the model can learn the background class information from the new data due to the repetition of many structural background classes in new data. Traditional replay-based methods store the original pixels of these background classes from old data, resulting in low memory efficiency. To enhance memory efficiency, we propose Spatial–Adaptive replay for Foreground objects (SAF), a method that stores only foreground-class pixels and their spatial information. In addition, Spatial–Adaptive Mix-up (SAM) is designed in the proposed method, based on the spatial distribution characteristics of the foreground classes. Spatial–adaptive alignment ensures that foreground objects are mixed with new samples in a reasonable spatial arrangement to help the model learn critical contextual information. Experiments on the Cityscapes and BDD100K datasets show that the proposed method obtains competitive results in class-incremental semantic segmentation tasks for autonomous driving scenarios.

  • Open Access

    Adaptive Shadow Compensation Method in Hyperspectral Images via Multi-Exposure Fusion and Edge Fusion

    by Yan Meng, Guanyi Li, Wei Huang

    Shadows in hyperspectral images lead to reduced spectral intensity and changes in spectral characteristics, significantly hindering analysis and applications. However, current shadow compensation methods face the issue of nonlinear attenuation at different wavelengths and unnatural transitions at the shadow boundary. To address these challenges, we propose a two-stage shadow compensation method based on multi-exposure fusion and edge fusion. Initially, shadow regions are identified through color space conversion and an adaptive threshold. The first stage utilizes multi-exposure, generating a series of exposure images through adaptive exposure coefficients that reflect spatial shadow intensity variations. Fusion weights for exposure images are determined based on exposure, contrast, and spectral variance. Then, the exposure sequence and fusion weights are constructed as Laplacian pyramids and Gaussian pyramids, respectively, to obtain a weighted fused exposure sequence. In the second stage, the previously identified shadow regions are smoothly reintegrated into the original image using edge fusion based on the p-Laplacian operator. To further validate the effectiveness and spectral fidelity of our method, we introduce a new hyperspectral image dataset. Experimental results on the public dataset and proposed dataset demonstrate that our method surpasses other mainstream shadow compensation methods.


  • Open Access

    Self-Supervised Non-Uniform Low-Light Image Enhancement Combining Image Inversion and Exposure Fusion

    by Wei Huang, Kaili Li, Mengfan Xu, Rui Huang

    Low-light image enhancement is a challenging task in non-uniform low-light conditions, often resulting in local overexposure, noise amplification, and color distortion. To obtain satisfactory enhancement results, most models must resort to carefully selected paired or multi-exposure data sets. In this paper, we propose a self-supervised framework for non-uniform low-light image enhancement to address these issues, only requiring low-light images on their own for training. We first design a robust Retinex model-based image exposure enhancement network (EENet) to obtain global brightness enhancement and noise removal of images by carefully designing the loss function of each decomposition map. Then, to correct overexposed areas in the enhanced image, we incorporate the inverse image of the low-light image for enhancement using EENet. Furthermore, a three-branch asymmetric exposure fusion network (TAFNet) is designed. The two enhanced images and the original image are used as the TAFNet inputs to obtain a globally well-exposed and detail-rich image. Experimental results demonstrate that our framework outperforms some state-of-the-art methods in visual and quantitative comparisons.


  • Open Access

    Near-Infrared Image Colorization Using Asymmetric Codec and Pixel-Level Fusion

    by Xiaoyu Ma, Wei Huang, Rui Huang, Xuefeng Liu

    This paper mainly studies the colorization of near-infrared (NIR) images. Image colorization methods cannot be extended to NIR image colorization since the wavelength band of the NIR image exceeds the visible light spectral range and it is often linearly independent of the luminance of the RGB image. Furthermore, a symmetric codec, which cannot guarantee the ability of the encoder to extract features, is often used as the main frame of the network in both CNN-based colorization networks and CycleGAN-based colorization networks. In order to deal with the investigated problem, we propose a novel NIR colorization method using asymmetric codec (ACD) and pixel-level fusion. ACD is designed to improve the feature extraction ability of the encoder by allowing the information to enter deeper into the model and learning more non-redundant information. In addition, the global and local feature fusion networks (GLFFNet) are embedded between the encoder and the decoder to improve the prediction of the subtle color information of the image. The ACD and GLFFNet together constitute the colorization network (ColorNet) in this paper. Bilateral filtering and weighted least squares filtering (BFWLS) are used to fuse the pixel-level information of the input NIR image into the raw output image of the ColorNet. Finally, an intensive comparison analysis based on common datasets is conducted to verify superiority over existing methods in qualitative and quantitative visual assessments.


  • A high fidelity haze removal algorithm for optical satellite images using progressive transmission estimation based on the dark channel prior

    by Wei Huang, Yueyun Wang, Rui Wang

    In this paper, a high fidelity haze removal algorithm is proposed for optical satellite images by using progressive transmission estimation based on the dark channel prior (DCP). The transmission is estimated adaptively according to the histogram of the dark channel image and the constraint of maximum transmission. Then, the guided filter is used to refine the transmission to obtain a continuous transmission map in which the clean areas are retained as much as possible. The refined transmission is applied to each visual band to obtain the initial de-hazing image. Then, the transmission is re-evaluated for the initial de-hazing image, and a guided filter with a small window size is used to refine the re-evaluated transmission. Furthermore, the transmission is stretched with the power-law transformation (PLT). To ensure fidelity in hazy areas, the optimal stretched transmission is estimated according to the artificially selected samples, from which the final haze removal results can be achieved. Several optical satellite images are collected and tested to validate the effectiveness of the proposed method. The evaluation results demonstrate that the proposed method is superior to the traditional methods and can recover a haze-free image with high fidelity.


  • Cloud detection for high-resolution remote-sensing images of urban areas using colour and edge features based on dual-colour models

    by Wei Huang, Yueyun Wang, Xiu Chen

    In this article, the colour and edge features of clouds are proposed based on the analysis of a large number of cloudy high-resolution remote-sensing images of urban area. Using the cloud features, a cloud detection method based on dual-colour space is proposed. First, two candidate cloud regions maps are built based on hue–saturation–intensity model and blue, green, red model, respectively. According to the superposition analysis of the two maps, fine cloud regions map are obtained. After removing the fine cloud regions from the image, the image is linearly stretched to enhance the intensity of thin clouds around the fine cloud regions. In the stretched image, the thin cloud regions map is built with the threshold technique. Finally, the fine cloud regions map and the thin cloud regions map are combined to form the complete cloud regions map. The results of experiments and comparisons indicate that our algorithm has good accuracy and efficiency.