Enhancing action recognition of construction workers using data-driven scene parsing

Abstract

Vision-based action recognition of construction workers has attracted increasing attention for its diverse applications. Though state-of-the-art performances have been achieved using spatial-temporal features in previous studies, considerable challenges remain in the context of cluttered and dynamic construction sites. Considering that workers actions are closely related to various construction entities, this paper proposes a novel system on enhancing action recognition using semantic information. A data-driven scene parsing method, named label transfer, is adopted to recognize construction entities in the entire scene. A probabilistic model of actions with context is established. Worker actions are first classified using dense trajectories, and then improved by construction object recognition. The experimental results on a comprehensive dataset show that the proposed system outperforms the baseline algorithm by 10.5%. The paper provides a new solution to integrate semantic information globally, other than conventional object detection, which can only depict local context. The proposed system is especially suitable for construction sites, where semantic information is rich from local objects to global surroundings. As compared to other methods using object detection to integrate context information, it is easy to implement, requiring no tedious training or parameter tuning, and is scalable to the number of recognizable objects.

Construction Activity Recognition Method Based on Object Detection, Attention Orientation Estimation, and Person Re-Identification

“…In early construction activity recognition research [46][47][48], traditional image processing methods are more commonly used. However, with the development and improvement of CNN, the training cost of the object detection algorithm has been further reduced, resulting in improved detection speed and accuracy.…”

Review of Optimization Dynamically Applied in the Construction and the Application Potential of ICT

“…Moreover, many other ICT-related studies can provide dynamic data to periodic mode. Through the recognition of workers' action [61], construction equipment action [62], or the monitoring of the hoisting status of the precast wall [63], the current construction process can be determined. By extracting both temporal and spatial information of the equipment from a long video, the current productivity can be analyzed [64,65].…”

Идентификация Поведенческих Паттернов

“…Поскольку не все эти преступления напрямую направлены против людей, в[8] на его основе был собран набор данных, содержащий подмножество преступлений, связанных с людьми.Также алгоритмы видеоаналитики могут быть использованы на промышленных предприятиях. В[9][10][11][12] представлены варианты практического применения распознавания действий на строительной площадке. Поскольку представленные наборы данных содержат типовые действия, происходящие на строительной площадке, анализ техники безопасности с применением этих наборов затруднителен, поскольку отсутствуют классы аномальных действий.Кроме того, анализ видеоданных может быть полезен в электроэнергетике.…”