• International conference on construction engineering and project management
• /
• 2020.12a
• /
• pp.313-322
• /
• 2020

The recognition of the risk hazards is a vital step to effectively prevent accidents on a construction site. The advanced development in computer vision systems and the availability of the large visual database related to construction site made it possible to take quick action in the event of human error and disaster situations that may occur during management supervision. Therefore, it is necessary to analyze the risk factors that need to be managed at the construction site and review appropriate and effective technical methods for each risk factor. This research focuses on analyzing Occupational Safety and Health Agency (OSHA) related to risk zone identification rules that can be adopted by the image recognition technology and classify their risk factors depending on the effective technical method. Therefore, this research developed a pattern-oriented classification of OSHA rules that can employ a large scale of safety hazard recognition. This research uses joint reasoning of risk zone Identification and numeric input by utilizing a stereo camera integrated with an image detection algorithm such as (YOLOv3) and Pyramid Stereo Matching Network (PSMNet). The research result identifies risk zones and raises alarm if a target object enters this zone. It also determines numerical information of a target, which recognizes the length, spacing, and angle of the target. Applying image detection joint logic algorithms might leverage the speed and accuracy of hazard detection due to merging more than one factor to prevent accidents in the job site.

• Computers and Concrete
• /
• v.33 no.5
• /
• pp.535-544
• /
• 2024

In the rapidly advancing landscape of computer vision (CV) technology, there is a burgeoning interest in its integration with the construction industry. Camera calibration is the process of deriving intrinsic and extrinsic parameters that affect when the coordinates of the 3D real world are projected onto the 2D plane, where the intrinsic parameters are internal factors of the camera, and extrinsic parameters are external factors such as the position and rotation of the camera. Camera pose estimation or extrinsic calibration, which estimates extrinsic parameters, is essential information for CV application at construction since it can be used for indoor navigation of construction robots and field monitoring by restoring depth information. Traditionally, camera pose estimation methods for cameras relied on target objects such as markers or patterns. However, these methods, which are marker- or pattern-based, are often time-consuming due to the requirement of installing a target object for estimation. As a solution to this challenge, this study introduces a novel framework that facilitates camera pose estimation using standardized materials found commonly in construction sites, such as concrete forms. The proposed framework obtains 3D real-world coordinates by referring to construction materials with certain specifications, extracts the 2D coordinates of the corresponding image plane through keypoint detection, and derives the camera's coordinate through the perspective-n-point (PnP) method which derives the extrinsic parameters by matching 3D and 2D coordinate pairs. This framework presents a substantial advancement as it streamlines the extrinsic calibration process, thereby potentially enhancing the efficiency of CV technology application and data collection at construction sites. This approach holds promise for expediting and optimizing various construction-related tasks by automating and simplifying the calibration procedure.

• Korean Journal of Cognitive Science
• /
• v.15 no.4
• /
• pp.1-14
• /
• 2004

When two objects are tying in the same visual direction there occurs abrupt depth change between two objects, which is against the assumption of the computational model for stereopsis on the surfaces in a natural scene. For this reason, this stimulus configuration is popularly used in the studies for the effectiveness of the constraints employed in the computational model. Contrary to the results from two nails (or objects) tying in the same visual direction, the two different surfaces from random-dot stereogram (RDS) in the same situation can be seen simultaneously in the different depth. The seemingly contradictory results between two situations my reflect the different strategies imposed by binocular mechanism for each situation during binocular matching process. Otherwise, the surfaces tying in the same visual direction is not equivalent situation to two objects tying in the same visual direction with regards to matching process. In order to examine above possibilities, the stereoscopic depth of the surface was measured after manipulating the visual direction of the surface elements. The visual direction of each dot pair from different surfaces in RDS (in Experiment 1) or the visual direction of line (hawing rectangle with regard to that of the vertical line (in Experiment 2) was manipulated. The stereoscopic depth of the surface was found to be varied depending on visual direction of the surface elements in both RDS and line hawing stimulus. Similar to the results from two nails situation depth of the surface was greatly reduced when each surface element was tying in the same visual direction as that of the other surface element or the other object. These results suggest that binocular mechanism imposes no different strategy in resolving correspondence problem in both two objects and two surfaces situation. And the results were discussed in the context of usefulness of the constraints employed in the computational model for stereopsis.

• Journal of KIISE:Software and Applications
• /
• v.31 no.1
• /
• pp.54-60
• /
• 2004

Face recognition is one of the problems to be solved by appearance based matching technique. However, the appearance of face image is very sensitive to variation in illumination. One of the easiest ways for better performance is to collect more training samples acquired under variable lightings but it is not practical in real world. ]:n object recognition, it is desirable to focus on feature extraction or normalization technique rather than focus on classifier. This paper presents a simple approach to normalization of faces subject to directional illumination. This is one of the significant issues that cause error in the face recognition process. The proposed method, ICR(illumination Compensation based on Multiple Linear Regression), is to find the plane that best fits the intensity distribution of the face image using the multiple linear regression, then use this plane to normalize the face image. The advantages of our method are simple and practical. The planar approximation of a face image is mathematically defined by the simple linear model. We provide experimental results to demonstrate the performance of the proposed ICR method on public face databases and our database. The experimental results show a significant improvement of the recognition accuracy.

• Journal of KIISE
• /
• v.44 no.8
• /
• pp.813-821
• /
• 2017

Entity Linking find the meaning of an entity mention, which indicate the entity using different expressions, in a user's query by linking the entity mention and the entity in the knowledge base. This task has four challenges, including the difficult knowledge base construction problem, multiple presentation of the entity mention, ambiguity of entity linking, and NIL entity recognition. In this paper, we first construct the entity name dictionary based on Wikipedia to build a knowledge base and solve the multiple presentation problem. We then propose various methods for NIL entity recognition and solve the ambiguity of entity linking by training the support vector machine based on several features, including the similarity of the context, semantic relevance, clue word score, named entity type similarity of the mansion, entity name matching score, and object popularity score. We sequentially use the proposed two methods based on the constructed knowledge base, to obtain the good performance in the entity linking. In the result of the experiment, our system achieved 83.66% and 90.81% F1 score, which is the performance of the NIL entity recognition to solve the ambiguity of the entity linking.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.4A
• /
• pp.239-249
• /
• 2012

Recently, the virtual view generation method using depth data is employed to support the advanced stereoscopic and auto-stereoscopic displays. Although depth data is invisible to user at 3D video rendering, its accuracy is very important since it determines the quality of generated virtual view image. Many works are related to such depth enhancement exploiting a time-of-flight (TOF) camera. In this paper, we propose a fast 3D scene capturing system using one TOF camera at center and two high-resolution cameras at both sides. Since we need two depth data for both color cameras, we obtain two views' depth data from the center using the 3D warping technique. Holes in warped depth maps are filled by referring to the surrounded background depth values. In order to reduce mismatches of object boundaries between the depth and color images, we used the joint bilateral filter on the warped depth data. Finally, using two color images and depth maps, we generated 10 additional intermediate images. To realize fast capturing system, we implemented the proposed system using multi-threading technique. Experimental results show that the proposed capturing system captured two viewpoints' color and depth videos in real-time and generated 10 additional views at 7 fps.

• Journal of Internet Computing and Services
• /
• v.18 no.1
• /
• pp.51-55
• /
• 2017

In this paper, we designed a LBS-based smart platform for Lifelog service that can utilize the other's lifelog information. The conventional Lifelog service means that the system records the daily activities of the smart device user so the user can retrieve the early-recorded information later. The proposed Lifelog service platform uses the GPS/UFID location information and the various information extracted from the image as the lifelog data. Further, the proposed Lifelog platform using DB can provide the user with the Lifelog data recorded by the other service user. The system usually provide the other's Lifelog data within the 500m distance from the user and the range of distance can be adjustable. The proposed smart platform based on image recognition for Lifelog can acquire the image from the smart device directly and perform the various image recognition processing to produce the useful image attributes. And it can store the location information, image data, image attributes and the relevant web informations on the database that can be retrieved by the other use's request. The attributes stored and managed in the image information database consist of the followings: Object ID, the image type, the capture time and the image GPS coordinates. The image type attribute has the following values: the mountain, the sea, the street, the front of building, the inside of building and the portrait. The captured image can be classified into the above image type by the pattern matching image processing techniques and the user's direct selection as well. In case of the portrait-attribute, we can choose the multiple sub-attribute values from the shirt, pant, dress and accessory sub-attributes. Managing the Lifelog data in the database, the system can provide the user with the useful additional services like a path finding to the location of the other service user's Lifelog data and information.

• Journal of Astronomy and Space Sciences
• /
• v.22 no.4
• /
• pp.385-392
• /
• 2005

We developed an algorithm to efficiently detect streaks in survey images and made a performance test with YSTAR-NEOPAT images obtained by the 0.5m telescope stationed in South Africa. Fast moving objects whose apparent speeds exceed 10 arcsec/min are the main target of our algorithm; these include artificial satellites, space debris, and very fast Near-Earth Objects. Our algorithm, based on the outline shape of elongated sources employs a step of image subtraction in order to reduce the confusion caused by dense distribution of faint stars. It takes less than a second to find and characterize streaks present in normal astronomical images of 2K format. Comparison with visual inspection proves the efficiency and completeness of our automated detection algorithm. When applied to about 7,000 time-series images from YSTAR telescope, nearly 700 incidents of streaks are detected. Fast moving objects are identified by the presence of matching streaks in adjoining frames. Nearly all of confirmed fast moving objects turn out to be artificial satellites or space debris. Majority of streaks are however meteors and cosmic ray hits, whose identity is often difficult to classify.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.5
• /
• pp.444-450
• /
• 2015

The accuracy of small and low cost CCD camera is insufficient to provide data for precisely tracking unmanned aerial vehicles(UAVs). This study shows how UAV can hover on a human targeted tracking object by using CCD camera rather than imprecise GPS data. To realize this, UAVs need to recognize their attitude and position in known environment as well as unknown environment. Moreover, it is necessary for their localization to occur naturally. It is desirable for an UAV to estimate of his attitude by environment recognition for UAV hovering, as one of the best important problems. In this paper, we describe a method for the attitude of an UAV using image information of a maker on the floor. This method combines the observed position from GPS sensors and the estimated attitude from the images captured by a fixed camera to estimate an UAV. Using the a priori known path of an UAV in the world coordinates and a perspective camera model, we derive the geometric constraint equations which represent the relation between image frame coordinates for a marker on the floor and the estimated UAV's attitude. Since the equations are based on the estimated position, the measurement error may exist all the time. The proposed method utilizes the error between the observed and estimated image coordinates to localize the UAV. The Kalman filter scheme is applied for this method. its performance is verified by the image processing results and the experiment.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.43 no.5 s.311
• /
• pp.29-37
• /
• 2006

Thresholding is a fundamental approach to segmentation that utilizes a significant degree of pixel popularity or intensity. Otsu's thresholding employed the normalized histogram as a discrete probability density function. Also it utilized a criterion that minimizes the between-class variance of pixel intensity to choose a threshold value for segmentation. However, the Otsu's method has a disadvantage of repeatedly searching optimal thresholds for the entire range. In this paper, a simple but fast multi-level thresholding approach is proposed by means of extending the Otsu's method. Rather than invoke the Otsu's method for the entire gray range, we advocate that the gray-level range of an image be first divided into smaller sub-ranges, and that the multi-level thresholds be achieved by iteratively invoking this dividing process. Initially, in the proposed method, the gray range of the object image is divided into 2 classes with a threshold value. Here, the threshold value for segmentation is selected by invoking the Otsu's method for the entire range. Following this, the two classes are divided into 4 classes again by applying the Otsu's method to each of the divided sub-ranges. This process is repeatedly performed until the required number of thresholds is obtained. Our experimental results for three benchmark images and fifty faces show a possibility that the proposed method could be used efficiently for pattern matching and face recognition.