ABSTRACT
Standardized usage of face masks is effective as a non-pharmaceutical intervention to prevent the spread of infectious respiratory diseases，such as COVID-19 and influenza. In the current epidemic situation，wearing face masks correctly is especially important. Most existing mask-wearing detection algorithms involve problems such as complex structures，high training difficulty，and insufficient feature extraction. Therefore，this study proposes a lightweight mask-wearing detection algorithm based on multi-scale feature fusion and the YOLOv4-Tiny network，called L-MFFN-YOLO. L-MFFN-YOLO improves on the original residual structure and uses a lightweight residual module to promote rapid convergence. Moreover，it reduces the computational load while ensuring detection accuracy. Based on the original network’s 13×13 and 26×26 feature maps，52×52 feature branches are added to enhance the ability of the lower feature layer to express information and reduce the false negative rate for small targets.On this basis，a Multi-level Cross Fusion （MCF） structure is used to maximally extract useful information so as to improve feature utilization. In addition to detecting mask-wearing，a category of masks worn incorrectly is added to the dataset and manually labeled. The www.eciexperimental results show that the size of the proposed L-MFFN-YOLO model is only 5.8 MB，which is 76% smaller than that of the original YOLOv4-Tiny. Moreover，the mean Average Precision（mAP）of the proposed approach is 5.25 percentage points higher，and its processing time is 14 ms faster on an equivalent CPU.These results demonstrate that the proposed approach can meet the requirements of accuracy and real-time operation in resource-constrained devices to detect faces wearing masks. © 2022, Editorial Office of Computer Engineering. All rights reserved.
ABSTRACT
BACKGROUND: The coronavirus disease 2019 (COVID-19) spreads rapidly across the globe, seriously threatening the health of people all over the world. To reduce the diagnostic pressure of front-line doctors, an accurate and automatic lesion segmentation method is highly desirable in clinic practice. PURPOSE: Many proposed two-dimensional (2D) methods for sliced-based lesion segmentation cannot take full advantage of spatial information in the three-dimensional (3D) volume data, resulting in limited segmentation performance. Three-dimensional methods can utilize the spatial information but suffer from long training time and slow convergence speed. To solve these problems, we propose an end-to-end hybrid-feature cross fusion network (HFCF-Net) to fuse the 2D and 3D features at three scales for the accurate segmentation of COVID-19 lesions. METHODS: The proposed HFCF-Net incorporates 2D and 3D subnets to extract features within and between slices effectively. Then the cross fusion module is designed to bridge 2D and 3D decoders at the same scale to fuse both types of features. The module consists of three cross fusion blocks, each of which contains a prior fusion path and a context fusion path to jointly learn better lesion representations. The former aims to explicitly provide the 3D subnet with lesion-related prior knowledge, and the latter utilizes the 3D context information as the attention guidance of the 2D subnet, which promotes the precise segmentation of the lesion regions. Furthermore, we explore an imbalance-robust adaptive learning loss function that includes image-level loss and pixel-level loss to tackle the problems caused by the apparent imbalance between the proportions of the lesion and non-lesion voxels, providing a learning strategy to dynamically adjust the learning focus between 2D and 3D branches during the training process for effective supervision. RESULT: Extensive experiments conducted on a publicly available dataset demonstrate that the proposed segmentation network significantly outperforms some state-of-the-art methods for the COVID-19 lesion segmentation, yielding a Dice similarity coefficient of 74.85%. The visual comparison of segmentation performance also proves the superiority of the proposed network in segmenting different-sized lesions. CONCLUSIONS: In this paper, we propose a novel HFCF-Net for rapid and accurate COVID-19 lesion segmentation from chest computed tomography volume data. It innovatively fuses hybrid features in a cross manner for lesion segmentation, aiming to utilize the advantages of 2D and 3D subnets to complement each other for enhancing the segmentation performance. Benefitting from the cross fusion mechanism, the proposed HFCF-Net can segment the lesions more accurately with the knowledge acquired from both subnets.