LZLiang Zhao

Liang Zhao

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Liang Zhao has a total of 19 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Confidence-Guided Fusion for Self-Supervised Monocular Depth Estimation in Endoscopy

    by Shuang Li, Hongbo Wang, Zhaoxu Hu, Tian Chu, Yingping Li, Liang Zhao

    Accurate monocular depth estimation (MDE) is a foundational task in endoscopic surgery, critical for augmenting depth perception and aiding surgical navigation. While diffusion-based and discriminative depth estimators demonstrate complementary strengths, they also exhibit asymmetric errors: discriminative models yield precise geometric boundaries but struggle in homogeneous or saturated areas, whereas diffusion models recover fine textures at the cost of occasional structural incoherence. To systematically exploit this complementarity, we present CoDepth, a novel framework that leverages confidence-guided fusion to harmonize the outputs of these heterogeneous estimators. Its core components include a complementary map extractor that identifies structured disparity disagreements, a cross-attention module for context-aware feature integration, and a probabilistic confidence network that generates spatially adaptive fusion weights. Extensive evaluations on the SCARED dataset show that CoDepth achieves improved overall performance relative to strong single-model baselines, with the most consistent gains observed in Abs Rel and δ-based accuracy, while changes in some other error metrics are more modest. Furthermore, CoDepth exhibits encouraging cross-domain generalization. When a model trained on SCARED is directly evaluated on SERV-CT, Hamlyn, and C3VD without fine-tuning, it achieves competitive performance and improves several key metrics across datasets. The framework also demonstrates enhanced robustness against common synthetic corruptions like low-light conditions, Gaussian noise, and impulse noise, underscoring its practical utility in complex clinical settings. These results suggest that confidence-guided complementary fusion provides a practical integration-level paradigm for combining heterogeneous endoscopic depth estimators.

  • Instance Segmentation and Length Quantification Analysis of Tunnel Face Joints

    by Liang Zhao, Hui Jing Li, Zhan Ping Song, Khasan Umarov

    Accurate segmentation and quantitative analysis of tunnel face joints are crucial for optimizing blasting parameters and ensuring construction safety. However, in complex tunnel environments, existing segmentation algorithms suffer from insufficient accuracy in segmenting small, elongated joint targets, leading to significant errors in subsequent length quantification. To address this issue, this paper proposes a segmentation algorithm U2Y11FL for the joint target of the tunnel face. The U2Y11FL algorithm utilizes UNetV2's encoder to extract basic features of the fractures, subsequently introduces the FDPN module to integrate feature information from different levels, and finally employs the LAWDS module for down-sampling. On a self-constructed joint dataset, the mask segmentation accuracy rate and Dice coefficient of U2Y11FL reach 85.4% and 72.9% respectively. On three open-source crack datasets, it demonstrated strong generalization capabilities with Dice coefficients as high as 85.6%, 79.4% and 76.5%. Based on this, a fractal approximation algorithm using chain codes was adopted to quantify the length characteristics of joints. By compensating for joint fractures occurring during the skeleton extraction stage in the original algorithm, the relative error between the quantified results and measured values was reduced to 2.2%~9.7%, providing an effective method for quantitative analysis of tunnel face joints.

  • Open Access

    Describe-to-Score: Text-Guided Efficient Image Complexity Assessment

    by Shipeng Liu, Zhonglin Zhang, Dengfeng Chen, Liang Zhao

    Accurately assessing image complexity (IC) is critical for computer vision, yet most existing methods rely solely on visual features and often neglect high-level semantic information, limiting their accuracy and generalization. We introduce vision-text fusion for IC modeling. This approach integrates visual and textual semantic features, increasing representational diversity. It also reduces the complexity of the hypothesis space, which enhances both accuracy and generalization in complexity assessment. We propose the D2S (Describe-to-Score) framework, which generates image captions with a pre-trained vision-language model. We propose the feature alignment and entropy distribution alignment mechanisms, D2S guides semantic information to inform complexity assessment while bridging the gap between vision and text modalities. D2S utilizes multi-modal information during training but requires only the vision branch during inference, thereby avoiding multi-modal computational overhead and enabling efficient assessment. Experimental results demonstrate that D2S outperforms existing methods on the IC9600 dataset and maintains competitiveness on no-reference image quality assessment (NR-IQA) benchmark, validating the effectiveness and efficiency of multi-modal fusion in complexity-related tasks. Code is available at: https://github.com/xauat-liushipeng/D2S

  • Open Access

    CLICv2: Image Complexity Representation via Content Invariance Contrastive Learning

    by Shipeng Liu, Liang Zhao, Dengfeng Chen

    Unsupervised image complexity representation often suffers from bias in positive sample selection and sensitivity to image content. We propose CLICv2, a contrastive learning framework that enforces content invariance for complexity representation. Unlike CLIC, which generates positive samples via cropping-introducing positive pairs bias-our shifted patchify method applies randomized directional shifts to image patches before contrastive learning. Patches at corresponding positions serve as positive pairs, ensuring content-invariant learning. Additionally, we propose patch-wise contrastive loss, which enhances local complexity representation while mitigating content interference. In order to further suppress the interference of image content, we introduce Masked Image Modeling as an auxiliary task, but we set its modeling objective as the entropy of masked patches, which recovers the entropy of the overall image by using the information of the unmasked patches, and then obtains the global complexity perception ability. Extensive experiments on IC9600 demonstrate that CLICv2 significantly outperforms existing unsupervised methods in PCC and SRCC, achieving content-invariant complexity representation without introducing positive pairs bias.

  • Open Access

    Contrastive Learning for Image Complexity Representation

    by Shipeng Liu, Liang Zhao, Dengfeng Chen, Zhanping Song

    Quantifying and evaluating image complexity can be instrumental in enhancing the performance of various computer vision tasks. Supervised learning can effectively learn image complexity features from well-annotated datasets. However, creating such datasets requires expensive manual annotation costs. The models may learn human subjective biases from it. In this work, we introduce the MoCo v2 framework. We utilize contrastive learning to represent image complexity, named CLIC (Contrastive Learning for Image Complexity). We find that there are complexity differences between different local regions of an image, and propose Random Crop and Mix (RCM), which can produce positive samples consisting of multi-scale local crops. RCM can also expand the train set and increase data diversity without introducing additional data. We conduct extensive experiments with CLIC, comparing it with both unsupervised and supervised methods. The results demonstrate that the performance of CLIC is comparable to that of state-of-the-art supervised methods. In addition, we establish the pipelines that can apply CLIC to computer vision tasks to effectively improve their performance.


  • Open Access

    Research on Path Planning of a Mining Inspection Robot in an Unstructured Environment Based on an Improved Rapidly Exploring Random Tree Algorithm

    by Jingwen Wu, Liang Zhao, Ruixue Liu

    To ensure the safe production of mines, the intelligent trend of underground mining operations is gradually advancing. However, the operational environment of subterranean mining is intricate, making the conventional path-planning algorithm used by mining inspection robots frequently inadequate for real requirements. To safeguard the mining inspection robot, targeting the problem of low search efficiency and path redundancy in the path planning of the existing rapidly exploring random tree (RRT) algorithm in the narrow and complex unstructured environment, a path-planning algorithm combining improved RRT and a probabilistic road map (PRM) is proposed. Initially, the target area is efficiently searched according to the fan-shaped goal orientation strategy and the adaptive step size expansion strategy. Subsequently, the PRM algorithm and the improved RRT algorithm are combined to reduce the redundant points of the planning path. Ultimately, considering the kinematics of the vehicle, the path is optimized by the third-order Bessel curve. The experimental simulation results show that the proposed path-planning algorithm has a higher success rate, smoother path, and shorter path length than other algorithms in complex underground mining environments, which proves the effectiveness of the proposed algorithm.


  • Open Access

    Research on Fault Detection by Flow Sequence for Industrial Internet of Things in Sewage Treatment Plant Case

    by Dongfeng Lei, Liang Zhao, Dengfeng Chen

    Classifying the flow subsequences of sensor networks is an effective way for fault detection in the Industrial Internet of Things (IIoT). Traditional fault detection algorithms identify exceptions by a single abnormal dataset and do not pay attention to the factors such as electromagnetic interference, network delay, sensor sample delay, and so on. This paper focuses on fault detection by continuous abnormal points. We proposed a fault detection algorithm within the module of sequence state generated by unsupervised learning (SSGBUL) and the module of integrated encoding sequence classification (IESC). Firstly, we built a network module based on unsupervised learning to encode the flow sequence of the different network cards in the IIoT gateway, and then combined the multiple code sequences into one integrated sequence. Next, we classified the integrated sequence by comparing the integrated sequence with the encoding fault type. The results obtained from the three IIoT datasets of a sewage treatment plant show that the accuracy of the SSGBUL–IESC algorithm exceeds 90% with subsequence length 10, which is significantly higher than the accuracies of the dynamic time warping (DTW) algorithm and the time series forest (TSF) algorithm. The proposed algorithm reaches the classification requirements for fault detection for the IIoT.


  • Open Access

    Short-Term Energy Consumption Prediction of Large Public Buildings Combined with Data Feature Engineering and Bilstm-Attention

    by Zeqin Tian, Dengfeng Chen, Liang Zhao

    Accurate building energy consumption prediction is a crucial condition for the sustainable development of building energy management systems. However, the highly nonlinear nature of data and complex influencing factors in the energy consumption of large public buildings often pose challenges in improving prediction accuracy. In this study, we propose a combined prediction model that combines signal decomposition, feature screening, and deep learning. First, we employ the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) to decompose energy consumption data. Next, we propose the Maximum Mutual Information Coefficient (MIC)-Fast Correlation Based Filter (FCBF) combined feature screening method for feature selection on the decomposed components. Finally, the selected input features and corresponding components are fed into the Bi-directional Long Short-Term Memory Attention Mechanism (BiLSTMAM) model for prediction, and the aggregated results yield the energy consumption forecast. The proposed approach is validated using energy consumption data from a large public building in Shaanxi Province, China. Compared with the other five comparison methods, the RMSE reduction of the CEEMDAN-MIC-FCBF-BiLSTMAM model proposed in this study ranged from 57.23% to 82.49%. Experimental results demonstrate that the combination of CEEMDAN, MIC-FCBF, and BiLSTMAM modeling markedly improves the accuracy of energy consumption predictions in buildings, offering a potent method for optimizing energy management and promoting sustainability in large-scale facilities.


  • Open Access

    Research on Fault Detection by Flow Sequence Algorithm with Deep Learning Model for Industrial Internet of Things

    by Dongfeng Lei, Liang Zhao, Dengfeng Chen

    Classifying network flow subsequences in an Industrial IoT gateway is an effective method for identifying Industrial IoT faults. This paper proposes a subsequence classification algorithm SSGBUL-IKNN based on unsupervised learning encoding to address the problem of low classification accuracy. Firstly, we built a network module based on unsupervised learning to encode the flow sequence of the Industrial IoT gateway. In the encoding module, a subsequence calibration algorithm controls the increasing error during the coding process. Then, the specific fault type is obtained through an improved KNN classification algorithm based on the indication distance source from the integrated coding sequences converted by multi-dimensional network flow sequences. The test result on three Industrial IoT datasets of a sewage treatment plant shows that the accuracy of SSGBUL-IKNN in this paper exceeds 92%, significantly higher than the algorithm DTW and TSF. It reaches the classification requirements of Industrial IoT fault diagnosis.

  • Open Access

    Research on Routing Algorithm of Construction Robot Cluster Enhanced Ad Hoc Network

    by Ningke Wei, Liang Zhao, Dengfeng Chen

    An enhanced self-assembling network routing algorithm is proposed for the problem of weak connectivity of communication networks caused by factors such as movement or environmental interference in the construction and operation, and the maintenance of construction robot clusters. Firstly, the dynamic forwarding probability is calculated based on the contribution of nodes joining routing paths to network connectivity, and the robust connectivity of the network is achieved by introducing the connectivity feedback mechanism; secondly, the influence of link quality evaluation index Q balanced hop count, residual energy, and load on link stability is used to select appropriate neighbors for nodes as the subsequent hop nodes; finally, the dynamic characteristics of nodes are combined with the topology control technology to eliminate low-quality links and optimize the topology by link maintenance time prediction and to set robot node priority. The simulation results show that the proposed algorithm can guarantee a network connectivity rate above 97% under heavy load, reduce the end-to-end delay, and improve the network survival time, providing a theoretical basis for achieving stable and reliable interconnection between building robot nodes.