Man-Woo Park

Automated vision tracking of project related entities

Tracking of project related entities such as construction equipment, materials, and personnel is used to calculate productivity, detect travel path conflicts, enhance the safety on the site, and monitor the project. Radio frequency tracking technologies (Wi-Fi, RFID, UWB) and GPS are commonly used for this purpose. However, on large-scale sites, deploying, maintaining and removing such systems can be costly and time-consuming. In addition, privacy issues with personnel tracking often limits the usability of these technologies on construction sites. This paper presents a vision based tracking framework that holds promise to address these limitations. The framework uses videos from a set of two or more static cameras placed on construction sites. In each camera view, the framework identifies and tracks construction entities providing 2D image coordinates across frames. Combining the 2D coordinates based on the installed camera system (the distance between the cameras and the view angles of them), 3D coordinates are calculated at each frame. The results of each step are presented to illustrate the feasibility of the framework.

Three-Dimensional Tracking of Construction Resources Using an On-Site Camera System

Vision trackers have been proposed as a promising alternative for tracking at large-scale, congested construction sites. They provide the location of a large number of entities in a camera view across frames. However, vision trackers provide only two-dimensional (2D) pixel coordinates, which are not adequate for construction applications. This paper proposes and validates a method that overcomes this limitation by employing stereo cameras and converting 2D pixel coordinates to three-dimensional (3D) metric coordinates. The proposed method consists of four steps: camera calibration, camera pose estimation, 2D tracking, and triangulation. Given that the method employs fixed, calibrated stereo cameras with a long baseline, appropriate algorithms are selected for each step. Once the first two steps reveal camera system parameters, the third step determines 2D pixel coordinates of entities in subsequent frames. The 2D coordinates are triangulated on the basis of the camera system parameters to obtain 3D coordinates. The methodology presented in this paper has been implemented and tested with data collected from a construction site. The results demonstrate the suitability of this method for on-site tracking purposes.

2D Vision Tracking Methods’ Performance Comparison for 3D Tracking of Construction Resources

In construction sites, tracking personnel, equipment, and materials is necessary in many applications such as asset management and progress monitoring. Vision tracking as a tracking technique has two unique capabilities which make it ideal for tracking on large scale, congested construction sites; tracking from a distance and tracking multiple elements from a single camera. 2D tracking results can be obtained through each camera and the 3D positions can be calculated by calibrating multiple cameras. 2D vision tracking methods vary and each method has unique capabilities, but little is known about the appropriateness of each method for tracking construction resources. This paper presents an evaluation of these methods that aims to identify the one most suitable for 3D tracking in construction sites. To satisfy this goal, 2D tracking methods are categorized, metrics are identified and performance of two categories of methods is evaluated based on these metrics. This evaluation has been used to find some preliminary results that can ultimately help with selection of the most suitable category of 2D vision tracking methods for construction applications.

Detection of Construction Workers in Video Frames for Automatic Initialization of Vision Trackers

Vision based tracking can provide the spatial location of construction entities such as equipment, workers, and materials in large scale, congested construction sites. It tracks entities in video streams by inferring their locations based on the entities’ visual features and motion histories. To initiate the process, it is necessary to determine the pixel areas corresponding to the construction entities to be tracked in the following consecutive video frames. In order to fully automate the process, an automated way of initialization is needed. This paper presents the method for construction worker detection which can automatically recognize and localize construction workers in video frames. The method first finds the foreground areas of moving objects using a background subtraction method. Within these foreground areas, construction workers are recognized based on the histogram of oriented gradients (HOG) and histogram of the HSV colors. HOG’s have proved to work effectively for detection of people, and the histogram of HSV colors helps differentiate between pedestrians and construction workers wearing safety vests. Preliminary experiments show that the proposed method has the potential to automate the initialization process of vision based tracking.

Automated Brick Counting for Façade Construction Progress Estimation

AbstractConstruction progress is predominantly measured with manual site surveys. These surveys are labor-intensive, on-site manual investigations. The generated survey reports are subjective and approximate because they are based on the surveyors’ individual experiences. This paper presents a novel method that can automatically count the number of bricks on a facade for reducing the cost and increasing the reliability of progress surveys. The method uses video data taken from a user’s mobile phone to detect bricks on a facade in each video frame by using color thresholding, edge detection, and filtering of rectangular shapes and sizes. Then, it compares the difference between consecutive frames to add counts when new bricks appear and to avoid double counting. The proposed method was implemented and tested on on-site videos of red brick facades, and resulted in 99.8% precision and 98.7% recall. The results demonstrate the suitability of this method for progress monitoring of brick facade construction.

Initializing Vision Based Trackers Using Semantic Texton Forests

Vision based tracking can provide the spatial location of project related entities such as equipment, workers, and materials in a large-scale congested construction site. It tracks entities in a video stream by inferring their motion. To initiate the process, it is required to determine the pixel areas of the entities to be tracked in the following consecutive video frames. For the purpose of fully automating the process, this paper presents an automated way of initializing trackers using Semantic Texton Forests (STFs) method. STFs method performs simultaneously the segmentation of the image and the classification of the segments based on the low-level semantic information and the context information. In this paper, STFs method is tested in the case of wheel loaders recognition. In the experiments, wheel loaders are further divided into several parts such as wheels and body parts to help learn the context information. The results show 79% accuracy of recognizing the pixel areas of the wheel loader. These results signify that STFs method has the potential to automate the initialization process of vision based tracking.

Automated In-Placed Brick Counting for Façade Construction Progress Estimation

© ASCE 2014. The prevalent method of measuring progress is through manual site surveys. These surveys are tedious and time-consuming. They are also approximate, as counting the number of bricks in-place to compare against those ordered is a very laborious task compared to its end value. In previous research, the authors were able to count the number of bricks using single images. This paper presents a novel method for counting bricks in-placed to automate the brick site survey using digital videos. This method improves the brick counting method on image to a continuous counting in video frames. It can compare the brick detection results in successive frames and accumulate the counts when new bricks appear. The method works by the following two steps: (1) count the number of bricks on the brick facade on a single video frame, and (2) track the detected bricks with a kernel-based tracking approach to compare the difference in successive video frames to avoid double counting, and add new bricks to the count upon the appearance in the successive video frames. This paper also demonstrated a performance comparison of the kernel-based tracking approach and point-based tracking approach. Test results demonstrate that this method is capable of counting the number of bricks on a brick facade with acceptable error.

Full-Body Occlusion Handling and Density Analysis in Traffic Video-Surveillance Systems

Vision-based traffic surveillance systems are among the most reliable, inexpensive, and highly applicable methodologies for surveying traffic conditions. The implementation of these strategies, however, is limited under certain conditions, such as the presence of vehicle occlusions or poor illumination conditions that lead to either overcounted or under-counted traffic data. The proposed motion-based methodology is intended to overcome these limitations by using a new technique for full-body occlusion handling of vehicles. The methodology is based on five main steps: calibration, detection, tracking, counting, and occlusion handling. The proposed methodology was tested with various 30–min videos and 452 preidentified cases of occlusion. Preliminary results indicated that the proposed methodology was reliable and robust in providing traffic density analysis. Future work may rely on the extension of the proposed methodology to deal with the detection of vehicles moving toward multiple directions.

Unsupervised Framework for Traffic Counting: Speed Estimation Based on Camera Network Data

A variety of traffic data, such as traffic counts and speed estimations, can be harvested from camera network systems installed along highways. This is possible through computer vision-based traffic monitoring processes that are mainly composed of vehicle detection and tracking, and field of view calibration. Several such processes have been proposed; however, they have not been fully validated on managing occlusion-based scenarios and generating reliable data over long periods of time and high volumes of traffic. This paper presents an effective, semi-automated method of detecting and tracking highway vehicles. The method integrates automated calibration of the field of view, detection and tracking. Trajectories, lanes, speeds and counts of tracked vehicles can be obtained from the videos using the proposed method. When a vehicle gets occluded by the other in adjacent lanes, the method identifies it based on the speed and acceleration, and terminates the tracking. When the vehicle reappears, it initiates a new tracking process. For validation, the framework is tested on videos recorded from CCTVs along the I-85 in Georgia, and evaluated on the accuracy of vehicle counting and speed. The tracked vehicles are counted when passing by pre-determined counting zones to avoid double counting. The speed results were compared with global positioning system (GPS) data. The results indicate that the proposed system has a potential to minimize human intervention and provide reliable counting and speed data.