Current semantic segmentation methods mainly focus on modeling the context of the global image to obtain high-quality segmentation results. However, they ignore the role of local image patches, which contain complementary and effective context information. In this paper, we propose an adaptive post-processing network (APPNet) for semantic segmentation based on the predictions of current methods in the global image and local image patches. The key point of APPNet is the global-local aggregation module, which models the context between global predictions and local predictions to generate accurate pixel-wise representation. Furthermore, we develop an adaptive points replacement module to compensate for the lack of fine detail in global prediction and the overconfidence in local predictions. Our method can be readily integrated into existing segmentation methods (i.e., ConvNeXt, HRNet, ViT-Adapter) with little memory and without extra modification in current models. We empirically demonstrate our method brings performance improvements across diverse datasets (i.e., Cityscapes, ADE20K, PASCAL-Context, COCO-Stuff). The code and models will be publicly available at https://github.com/zhu-gl-ux/APPN.

Class incremental semantic segmentation (CISS) aims to segment new classes during continual steps while preventing the forgetting of old knowledge. Existing methods alleviate catastrophic forgetting by replaying distributions of previously learned classes using stored prototypes or features. However, they overlook a critical issue: in CISS, the representation of class knowledge is updated continuously through incremental learning, whereas prototype replay methods maintain fixed prototypes. This mismatch between updated representation and fixed prototypes limits the effectiveness of the prototype replay strategy. To address this issue, we propose the Adaptive prototype replay (Adapter) for CISS in this paper. Adapter comprises an adaptive deviation compensation (ADC) strategy and an uncertainty-aware constraint (UAC) loss. Specifically, the ADC strategy dynamically updates the stored prototypes based on the estimated representation shift distance to match the updated representation of old class. The UAC loss reduces prediction uncertainty, aggregating discriminative features to aid in generating compact prototypes. Additionally, we introduce a compensation-based prototype similarity discriminative (CPD) loss to ensure adequate differentiation between similar prototypes, thereby enhancing the efficiency of the adaptive prototype replay strategy. Extensive experiments on Pascal VOC and ADE20K datasets demonstrate that Adapter achieves state-of-the-art results and proves effective across various CISS tasks, particularly in challenging multi-step scenarios.

Although promising results have been achieved in the areas of traffic-sign detection, little attention has paid to text detection on traffic signs. In fact, in today’s popular driver-less automobile industry, traffic sign text which brings abundant valuable traffic information plays an important and indispensable role. In this work, we design an effective scene text detector for traffic signs from street-level imagery, whose pipeline only consists of a preprocessing module to tackle with some complex situations, and a Scale-transfer layer in a Fully Convolutional Network (FCN) is proposed to speed up the network, and a simple post-processing step. Extensive experimental results show that the proposed method achieves the F-measure 0.82 on the Chinese traffic sign text dataset (CTST-1600), and the state-of-the-art experiment results on the ICDAR 2013 and MSRA-TD500, which show the robustness of the proposed method and its superiority to alternatives. We collect the text-based traffic signs dataset named CTST-1600 can be download at https://github.com/pummi823/test/blob/master/CTST-1600.

Traffic text detection is crucial for traffic scene understanding in intelligent transportation systems (ITS). Although natural scene text detection has been extensively studied, yielding noteworthy results, little research has focused on traffic text detection. Traffic text, as a special type of natural scene text, faces not only the general challenges of natural scene text detection but also the significant impact of false alarms from non-traffic texts on system performance. In light of these challenges, we propose an end-to-end traffic text detection framework that can effectively detect traffic text captured by in-vehicle cameras in various driving scenarios. The key contributions of our proposed approach are: (1) an Image Enhancement Module designed to remove fog and enhance low-quality images; (2) a plug-and-play Text Feature Enhancement Module; (3) a Joint Loss Function; and (4) the creation of an open driving environment traffic text dataset (named ODETT-3000) containing various traffic environments. Comprehensive experimental studies have verified that our method achieves state-of-the-art performance on traffic text datasets such as CTST-1600, TPD, and ODETT-3000, as well as promising results on public natural scene text datasets such as MSRA-TD500, ICDAR 2015, and CTW 1500, thereby showcasing the superior performance and adaptability of our method. The code and our dataset (ODETT-3000) will be available at https://github.com/yanbin-zhu/TTDNet.

Ultrasonography is a widely used medical imaging technique for detecting breast cancer. While manual diagnostic methods are subject to variability and time-consuming, computer-aided diagnostic (CAD) methods have proven to be more efficient. However, current CAD approaches neglect the impact of noise and artifacts on the accuracy of image analysis. To enhance the precision of breast ultrasound image analysis for identifying tissues, organs and lesions, we propose a novel approach for improved tumor classification through a dual-input model and global average pooling (GAP)-guided attention loss function. Our approach leverages a convolutional neural network with transformer architecture and modifies the single-input model for dual-input. This technique employs a fusion module and GAP operation-guided attention loss function simultaneously to supervise the extraction of effective features from the target region and mitigate the effect of information loss or redundancy on misclassification. Our proposed method has three key features: (i) ResNet and MobileViT are combined to enhance local and global information extraction. In addition, a dual-input channel is designed to include both attention images and original breast ultrasound images, mitigating the impact of noise and artifacts in ultrasound images. (ii) A fusion module and GAP operation-guided attention loss function are proposed to improve the fusion of dual-channel feature information, as well as supervise and constrain the weight of the attention mechanism on the fused focus region. (iii) Using the collected uterine fibroid ultrasound dataset to train ResNet18 and load the pre-trained weights, our experiments on the BUSI and BUSC public datasets demonstrate that the proposed method outperforms some state-of-the-art methods. The code will be publicly released at https://github.com/425877/Improved-Breast-Ultrasound-Tumor-Classification.

Identification of denatured biological tissue is crucial to high-intensity focused ultrasound (HIFU) treatment, which can monitor HIFU treatment and improve treatment efficiency. In this paper, a novel method based on compressed sensing (CS) and improved multiscale dispersion entropy (IMDE) is proposed to evaluate the complexity of ultrasonic scattered echo signals during HIFU treatment. In the analysis of CS, the method of orthogonal matching pursuit (OMP) is employed to reconstruct the denoised signal. CS-OMP can denoise the ultrasonic scattered echo signal effectively. Comparing with traditional multiscale dispersion entropy (MDE), IMDE improves the coarse-grained process in the multiscale analysis, which improves the stability of MDE. In the analysis of simulated signals, the entropy value of the IMDE method has less fluctuation compared with MDE, indicating that the IMDE method has better stability. In addition, MDE and IMDE are applied to the 300 cases of ultrasonic scattered echo signals after denoising (including 150 cases of normal tissues and 150 cases of denatured tissues). The experimental results show that the MDE and IMDE values of denatured tissues are higher than normal tissues. Both the MDE and IMDE method can be used to identify whether biological tissue is denatured. However, the multiscale entropy curve of IMDE is smoother and more stable than MDE. The interclass distance of IMDE is greater than MDE, and the intraclass distance of IMDE is less than MDE at different scale factors. This indicates that IMDE can better distinguish normal tissues and denatured tissues to obtain more accurate clinical diagnosis during HIFU treatment.