Jidong J. Yang’s research while affiliated with University of Georgia and other places

Image deraining holds great potential for enhancing the vision of autonomous vehicles in rainy conditions, contributing to safer driving. Previous works have primarily focused on employing a single network architecture to generate derained images. However, they often fail to fully exploit the rich prior knowledge embedded in the scenes. Particularly, most methods overlook the depth information that can provide valuable context about scene geometry and guide more robust deraining. In this work, we introduce a novel learning framework that integrates multiple networks: an AutoEncoder for deraining, an auxiliary network to incorporate depth information, and two supervision networks to enforce feature consistency between rainy and clear scenes. This multi-network design enables our model to effectively capture the underlying scene structure, producing clearer and more accurately derained images, leading to improved object detection for autonomous vehicles. Extensive experiments on three widely used datasets demonstrated the effectiveness of our proposed method.

Deep learning-based object detection models perform well under daytime conditions but face significant challenges at night, primarily because they are predominantly trained on daytime images. Additionally, training with nighttime images presents another challenge: Even human annotators struggle to accurately label objects in low-light conditions. This issue is particularly pronounced in transportation applications, such as detecting vehicles and other objects of interest on rural roads at night, where street lighting is often absent, and headlights may introduce undesirable glare. In this study, we addressed these challenges by introducing a novel framework for labeling-free data augmentation, leveraging synthetic data generated by the Car Learning to Act (CARLA) simulator for day-to-night image style transfer. Specifically, the framework incorporated the efficient attention Generative Adversarial Network for realistic day-to-night style transfer and used CARLA-generated synthetic nighttime images to help the model learn the vehicle headlight effect. To evaluate the efficacy of the proposed framework, we fine-tuned the state-of-the-art object detection model with an augmented dataset curated for rural nighttime environments, achieving significant improvements in nighttime vehicle detection. This novel approach was simple yet effective, offering a scalable solution to enhance deep learning-based detection systems in low-visibility environments and extended the applicability of object detection models to broader real-world contexts.

Autonomous vehicles face significant challenges in navigating adverse weather, particularly rain, due to the visual impairment of camera-based systems. In this study, we leveraged contemporary deep learning techniques to mitigate these challenges, aiming to develop a vision model that processes live vehicle camera feeds to eliminate rain-induced visual hindrances, yielding visuals closely resembling clear, rain-free scenes. Using the Car Learning to Act (CARLA) simulation environment, we generated a comprehensive dataset of clear and rainy images for model training and testing. In our model, we employed a classic encoder-decoder architecture with skip connections and concatenation operations. It was trained using novel batching schemes designed to effectively distinguish high-frequency rain patterns from low-frequency scene features across successive image frames. To evaluate the model performance, we integrated it with a steering module that processes front-view images as input. The results demonstrated notable improvements in steering accuracy, underscoring the model's potential to enhance navigation safety and reliability in rainy weather conditions.

Image deraining holds great potential for enhancing the vision of autonomous vehicles in rainy conditions, contributing to safer driving. Previous works have primarily focused on employing a single network architecture to generate derained images. However, they often fail to fully exploit the rich prior knowledge embedded in the scenes. Particularly, most methods overlook the depth information that can provide valuable context about scene geometry and guide more robust deraining. In this work, we introduce a novel learning framework that integrates multiple networks: an AutoEncoder for deraining, an auxiliary network to incorporate depth information, and two supervision networks to enforce feature consistency between rainy and clear scenes. This multi-network design enables our model to effectively capture the underlying scene structure, producing clearer and more accurately derained images, leading to improved object detection for autonomous vehicles. Extensive experiments on three widely-used datasets demonstrated the effectiveness of our proposed method.

Existing deep learning-based object detection models perform well under daytime conditions but face significant challenges at night, primarily because they are predominantly trained on daytime images. Additionally, training with nighttime images presents another challenge: even human annotators struggle to accurately label objects in low-light conditions. This issue is particularly pronounced in transportation applications, such as detecting vehicles and other objects of interest on rural roads at night, where street lighting is often absent, and headlights may introduce undesirable glare. This study addresses these challenges by introducing a novel framework for labeling-free data augmentation, leveraging CARLA-generated synthetic data for day-to-night image style transfer. Specifically, the framework incorporates the Efficient Attention Generative Adversarial Network for realistic day-to-night style transfer and uses CARLA-generated synthetic nighttime images to help the model learn vehicle headlight effects. To evaluate the efficacy of the proposed framework, we fine-tuned the YOLO11 model with an augmented dataset specifically curated for rural nighttime environments, achieving significant improvements in nighttime vehicle detection. This novel approach is simple yet effective, offering a scalable solution to enhance AI-based detection systems in low-visibility environments and extend the applicability of object detection models to broader real-world contexts.

Reliable and interpretable traffic crash modeling is essential for understanding causality and improving road safety. This study introduces a novel approach to predicting collision types by utilizing a comprehensive dataset fused from multiple sources, including weather data, crash reports, high-resolution traffic information, pavement geometry, and facility characteristics. Central to our approach is the development of a Feature Group Tabular Transformer (FGTT) model, which organizes disparate data into meaningful feature groups, represented as tokens. These group-based tokens serve as rich semantic components, enabling effective identification of collision patterns and interpretation of causal mechanisms. The FGTT model is benchmarked against widely used tree ensemble models, including Random Forest, XGBoost, and CatBoost, demonstrating superior predictive performance. Furthermore, model interpretation reveals key influential factors, providing fresh insights into the underlying causality of distinct crash types.

Traffic sensors are vital to the development and operation of Intelligent Transportation Systems, providing essential data for traffic monitoring, management, and transportation infrastructure planning. However, optimizing the placement of these sensors, particularly across large and complex statewide highway networks, remains a challenging task. In this research, we presented a novel search algorithm designed to address this challenge by leveraging information gradients from K-nearest neighbors within an embedding space. Our method enabled more informed and strategic sensor placement under budget and resource constraints, enhancing overall network coverage and data quality. Additionally, we incorporated spatial kriging analysis, harnessing spatial correlations of existing sensors to refine and reduce the search space. Our proposed approach was tested against the widely used Genetic Algorithm, demonstrating superior efficiency in terms of convergence time and producing more effective solutions with reduced information loss.

Metropolitan traffic networks are becoming increasingly complex due to the growing population and diverse range of travel modes. However, the limited installation of continuous count stations leads to partially observable networks, posing a significant challenge for effective highway planning and design practices at various scales. Travel demand models have been developed and calibrated using sparse traffic counts at the metropolitan level. Nevertheless, these models are cumbersome to recalibrate and rerun whenever network changes occur. To overcome this challenge, we propose a flexible learning-based approach that extracts embedded knowledge from large-scale activity-based travel demand models to estimate Annual Average Daily Traffic (AADT). The approach offers two primary advantages: (1) directly learning network flow patterns based on segment attributes and network topology that can be transferred across regions, and (2) enabling efficient and reliable AADT estimation for projects of various scales. Our study explores a wide range of machine learning techniques, including novel graph neural networks that explicitly account for network topology, as well as modern and traditional regression and regression kriging models, which either disregard or implicitly consider network topology. We conducted extensive experiments using the loaded network data from the activity-based travel demand model for the Atlanta metropolitan area. Our findings underscore the importance of network topology in AADT estimation, with the diffusion graph convolutional network model demonstrating the best performance in both transductive and inductive settings. Additionally, modern tree ensemble models such as random forest regressor and CatBoost, despite their ignorance of network topology, show the second-best inductive performance with relatively lightweight structures.