Accurate segmentation of fine-grained information is an important step in medical image analysis applications. With the development of the encoder-decoder-based networks, various network structures and algorithms have made significant progress in semantic segmentation tasks. This work aims to present a novel high-resolution encoder-decoder network (HRED-Net) for fine-grained image segmentation that is highly accurate for small-scale targets. We design a multiscale context connection module to extract feature information without reducing the resolution, and propose a multiresolution fusion model to fine-tune the final results. In addition, these modules are trained together with a detail-oriented loss function to enhance the model's perception of fine-grained parts. Through experiments on the DRIVE dataset, we found a balance between these modules, and our comparison results show that in addition to the extraction multiscale features, the fusion of multiresolution prediction information is also beneficial for fine-grained segmentation. Our method yielded significant improvements in the accuracy and sensitivity in retinal vessel and lung segmentation tasks.
In this paper, we present a novel framework for interactive segmentation problems. It integrates a nonparametric model into the Conditional Random Field (CRF) framework, which can effectively combine high-level features with low-level features to represent image information. In the nonparametric model, multiple region layers are used to estimate data likelihood terms to overcome the bad regions generated by unsupervised methods. The likelihood values of each layer are calculated separately to reduce the computational cost. In addition, we analyze that the pixel layer has little effect on data estimation, so we remove it to further reduce the complexity of the algorithm. We employ the label consistency between pixels and their corresponding regions in smooth term estimation, which can be regard as a higher order potential for pixels. The data term and the smooth term are then performed together in Conditional Random Fields (CRFs) as a fine-tuning of the results. Experimental results show that the proposed method can segment images efficiently and accurately with fewer user inputs.
Data-driven semantic segmentation models often generalize poorly to unseen adverse weather conditions due to significant domain shifts and the lack of labeled data in the target domain. While recent unsupervised domain adaptation (UDA) approaches attempt to mitigate this gap through adversarial alignment or self-training, they typically treat domain discrepancy as a holistic shift, leading to suboptimal adaptation under diverse and complex weather conditions. We propose a Dual-Path Domain Bridging with Atmospheric Correction (DPAC) framework that disentangles style-related and weather-related discrepancies via two complementary teacher models. First, a reference domain is introduced to decompose the source–target gap, and the teacher models are trained using dual-path mixed data to mitigate discrepancies caused by style variation and weather degradation. Then, an atmospheric correction module is integrated into the weather-path teacher to restore degraded target images and improve feature-level alignment. Lastly, a student model is optimized under the alternating supervision of both teachers, enabling robust knowledge integration and enhanced generalization. Extensive experiments on public benchmarks demonstrate the superiority of our approach. DPAC achieves significant performance gains of 5.1% on the Cityscapes to ACDC task, 6.1% on Cityscapes to Foggy Driving, and 5.3% on Cityscapes to Foggy Zurich over a strong baseline, validating the effectiveness of the dual-path domain bridging strategy in complex multi-weather scenarios.
1 January 2024 in International Journal of New Developments in Education
The course 'Principles and Applications of Microcontrollers' has a significant gap between theoretical knowledge and practical application, lacking real-world case studies. The content updates lag behind technological advancements, resulting in outdated knowledge that fails to meet the needs of actual development. To improve teaching effectiveness, a cross-disciplinary, project-oriented curriculum framework using the CDIO (Conceive-Design-Implement-Operate) project-based teaching approach has been proposed. We integrate local characteristics and implement multidimensional teaching resource integration. A project-based learning path has been established based on the OBE-CDIO engineering education philosophy and a collaborative education model between schools and enterprises, jointly building and sharing high-quality teaching resources has been constructed. Focusing on automotive microcontrollers, we have perfected the curriculum system, covering automotive microcontroller simulation experiments, student competitions, and innovation and entrepreneurship projects.This innovative reform has benefited students from various engineering disciplines, significantly enhancing their innovative and practical skills. Students have showcased innovative ideas and addressed social issues in projects, achieving excellent results in related competitions within the field.
Interactive image segmentation has gained much popularity since it can enhance the segmentation performance with human interactions. In this paper, we propose a novel framework that can effectively estimate the binary labels based on Conditional Random Field (CRF) models combining region-level information with pixels-level information. The user scribbles are treated as soft constraints. We use pixel-layer with local CRF and region-layer with fully connected CRF to model the graph. Then we combine the pixel information and region information together to calculate the pixels' likelihood term. In contrast to using relations between the pixel-layer and the region-layer during the probability computation, we calculate the probabilities of pixels and regions separately to reduce computational complexity. After the likelihood computation, a graph cut based model is used to refine the pixel-level label predictions of the image in order to avoid label inconsistency in regions and capture the boundary details. Due to the less relationship between layers and utilizing the refinement, the proposed model can provide high accuracy and user inputs insensitive results. Our experiments show that the proposed approach can efficiently and accurately segment the images compared with other state of art methods with only a few scribbles. Graph cut based fine tuning improves our segmentation accuracy and our method needs less computational cost than traditional multi-layer-based methods.