Pneumonia is a severe inflammation of the lung that could cause serious complications. Chest X-rays (CXRs) are commonly used to make a diagnosis of pneumonia. In this paper, we propose a deep-learning-based method with spatial attention superposition (SAS) and multilayer feature fusion (MFF) to facilitate pneumonia diagnosis based on CXRs. Specifically, an SAS module, which takes advantage of the channel and spatial attention mechanisms, was designed to identify intrinsic imaging features of pneumonia-related lesions and their locations, and an MFF module was designed to harmonize disparate features from different channels and emphasize important information. These two modules were concatenated to extract critical image features serving as the basis for pneumonia diagnosis. We further embedded the proposed modules into a baseline neural network and developed a model called SAS-MFF-YOLO to diagnose pneumonia. To validate the effectiveness of our model, extensive experiments were conducted on two CXR datasets provided by the Radiological Society of North America (RSNA) and the AI Research Institute. SAS-MFF-YOLO achieved a precision of 88.1%, a recall of 98.2% for pneumonia classification and an AP${}_{50}$ of 99% for lesion detection on the AI Research Institute dataset. The visualization of intermediate feature maps showed that our method could facilitate uncovering pneumonia-related lesions in CXRs. Our results demonstrated that our approach could be used to enhance the performance of the overall pneumonia detection on CXR imaging.

The automatic detection technology of helmet wearing is of great significance to the safety monitoring of construction site. It is difficult to detect a helmet when the object is small, overlapped, or unclear. Therefore, we propose SPZ-Det model to improve the performance of detect small targets in helmet detection task. We redesign the backbone network layers and proposes a Single Pixel Zoom (SPZ) module. SPZ-Det introduces the lower-level feature map, while for the same scale feature map, we select the higher-level feature map. Then the important elements of each feature map layer are amplified by SPZ module to solve the problem of small object feature fading. In the attention module of SPZ, the residual mechanism is introduced to weaken the influence of the attention. The proposed model improves the detection accuracy and keeps the detection speed. On the Safety-Helmet-Wearing dataset, the mAP of SPZ-Det is 27.9% higher than Efficientdet-d0 and gets an 8.8% increase compares with YOLOv3.