Partly due to the use of exhaustive-annotated data, deep networks have achieved impressive performance on medical image segmentation. Medical imaging data paired with noisy annotation are, however, ubiquitous, but little is known about the effect of noisy annotation on deep learning based medical image segmentation. We studied the effect of noisy annotation in the context of mandible segmentation from CT images. First, 202 images of head and neck cancer patients were collected from our clinical database, where the organs-at-risk were annotated by one of twelve planning dosimetrists. The mandibles were roughly annotated as the planning avoiding structure. Then, mandible labels were checked and corrected by a head and neck specialist to get the reference standard. At last, by varying the ratios of noisy labels in the training set, deep networks were trained and tested for mandible segmentation. The trained models were further tested on other two public datasets. Experimental results indicated that the network trained with noisy labels had worse segmentation than that trained with reference standard, and in general, fewer noisy labels led to better performance. When using 20% or less noisy cases for training, no significant difference was found on the segmentation results between the models trained by noisy or reference annotation. Cross-dataset validation results verified that the models trained with noisy data achieved competitive performance to that trained with reference standard. This study suggests that the involved network is robust to noisy annotation to some extent in mandible segmentation from CT images. It also highlights the importance of labeling quality in deep learning. In the future work, extra attention should be paid to how to utilize a small number of reference standard samples to improve the performance of deep learning with noisy annotation.

Recently, deep learning-based methods for synthetic aperture radar (SAR) ship detection have made remarkable advancements. However, most existing methods primarily focus on achieving high detection accuracy by employing complex models, leading to an increase in computational costs. In addition, some methods do not adequately consider the impact of speckle noise interference. To address these challenges, we propose a multiscale ship detector, called visual saliency-lightweight ship detector (VS-LSDet), utilizing visual saliency and lightweight convolutional neural network. First, a visual saliency enhancement module is proposed as a preprocessing step to visually highlight the ships and weaken the impact of speckle noise in the image. Second, a lightweight backbone called ghost-shuffle net (GSNet) is designed. We introduce two types of ghost-shuffle blocks as basic convolution blocks by introducing ghost convolution to reduce the model complexity and channel shuffle operation to enhance the representation ability of the feature map. Then, we propose a multishape dilated convolution block incorporated into GSNet to enlarge its receptive fields, further improving the detector's performance. Finally, a hybrid attention module (HyAM) is proposed, it leverages both spatial and channel information within the feature map. HyAM can emphasize ship-related features while suppressing irrelevant features from the background in the feature map. Experimental results on public SAR ship datasets demonstrate that, compared to other ship detection methods, VS-LSDet achieves higher detection accuracy with lower model complexity. Specifically, on the SSDD dataset, the AP value of VS-LSDet is 97.51%, with 2.53 M parameters and 6.21 GFLOPs.

Object detection in UAV images is an important and challenging task for many applications, which often needs highly efficient detection algorithms to meet the accuracy and real-time requirements of the applications. In this paper, we investigate efficient mechanisms for detecting dense and small objects in UAV images. Specifically, 1) kernel K-means is used to obtain optimal anchors for dense and small object detection; 2) a spatial information enhancement module (SIE) is proposed to improve the detection accuracy of dense objects by extracting object spatial location information; 3) a Coord_C3 module is proposed to improve the receptive field of the network and to reduce the number of network parameters; 4) a small detection head is added in the Head of network and skip connections are employed in the Neck of network to improve the detection accuracy of small objects. Experimental results on the VisDrone2019, LEVIR-ship and Stanford Drone datasets show that our method not only has higher detection accuracy, but also runs faster compared to state-of-the-art detection methods.

Water body segmentation is an important tool for the hydrological monitoring of the Earth. With the rapid development of convolutional neural networks, semantic segmentation techniques have been used on remote sensing images to extract water bodies. However, some difficulties need to be overcome to achieve good results in water body segmentation, such as complex background, huge scale, water connectivity, and rough edges. In this study, a water body segmentation model (DUPnet) with dense connectivity and multi-scale pyramidal pools is proposed to rapidly and accurately extract water bodies from Gaofen satellite and Landsat 8 OLI (Operational Land Imager) images. The proposed method includes three parts: (1) a multi-scale spatial pyramid pooling module (MSPP) is introduced to combine shallow and deep features for small water bodies and to compensate for the feature loss caused by the sampling process; (2) dense blocks are used to extract more spatial features to DUPnet’s backbone, increasing feature propagation and reuse; (3) a regression loss function is proposed to train the network to deal with the unbalanced dataset caused by small water bodies. The experimental results show that the F1, MIoU, and FWIoU of DUPnet on the 2020 Gaofen dataset are 97.67%, 88.17%, and 93.52%, respectively, and on the Landsat River dataset, they are 96.52%, 84.72%, 91.77%, respectively.

Speech emotion recognition (SER) aims to recognize human emotions through in-depth analysis of audio signals. However, it remains challenging to encode emotional cues and to fuse the encoded cues effectively. In this study, dual-stream representation is developed, and both full training and fine-tuning of different deep networks are employed for encoding emotion patterns. Specifically, a cross-attention fusion (CAF) module is designed to integrate the dual-stream output for emotion recognition. Using different dual-stream encoders (fully training a text processing network and fine-tuning a pre-trained large language network), the CAF module is compared to other three fusion modules on three databases. The SER performance is quantified with weighted accuracy (WA), unweighted accuracy (UA), and F1-score (F1S). The experimental results suggest that the CAF outperforms the other three modules and leads to promising performance on the databases (EmoDB: WA, 97.20%; UA, 97.21%; F1S, 0.8804; IEMOCAP: WA, 69.65%; UA, 70.88%; F1S, 0.7084; RAVDESS: WA, 81.86%; UA, 82.75.21%; F1S, 0.8284). It is also found that fine-tuning a pre-trained large language network achieves superior representation than fully training a text processing network. In a future study, improved SER performance could be achieved through the development of a multi-stream representation of emotional cues and the incorporation of a multi-branch fusion mechanism for emotion recognition.