Underground station passenger flow is large, the number of parcels carried by passengers is large and varied, and the parcels carried have an impact on the fire hazard and evacuation of the station. In order to determine the weights of the passenger luggage risk and environmental factor index system in the fire risk evaluation of underground stations in a more realistic way, an optimized and improved hierarchical analysis method for determining the judgement matrix is proposed, which improves the traditional nine-scaled method and adopts the three-scaled method for the four major categories of luggage, namely, handbags, rucksacks, portable power tools and trolley cases. The advantage of this method is that there is no need for consistency judgement in determining packages with a wide range of types and uncertain contents, thus simplifying the calculation. Meanwhile, the reasonableness and reliability of the method is verified by combining it with an actual metro station fire risk assessment system.

Falling is one of the most critical outcomes of loss of consciousness during triage in emergency department (ED). It is an important sign requires an immediate medical intervention. This paper presents a computer vision-based fall detection model in ED. In this study, we hypothesis that the proposed vision-based triage fall detection model provides accuracy equal to traditional triage system (TTS) conducted by the nursing team. Thus, to build the proposed model, we use MoveNet, a pose estimation model that can identify joints related to falls, consisting of 17 key points. To test the hypothesis, we conducted two experiments: In the deep learning (DL) model we used the complete feature consisting of 17 keypoints which was passed to the triage fall detection model and was built using Artificial Neural Network (ANN). In the second model we use dimensionality reduction Feature-Reduction for Fall model (FRF), Random Forest (RF) feature selection analysis to filter the key points triage fall classifier. We tested the performance of the two models using a dataset consisting of many images for real-world scenarios classified into two classes: Fall and Not fall. We split the dataset into 80% for training and 20% for validation. The models in these experiments were trained to obtain the results and compare them with the reference model. To test the effectiveness of the model, a t-test was performed to evaluate the null hypothesis for both experiments. The results show FRF outperforms DL model, and FRF has same accuracy of TTS.