• 한국멀티미디어학회논문지
• /
• 제7권12호
• /
• pp.1692-1699
• /
• 2004

Many researchers have developed various techniques for detection of human faces in ordinary still images. Face detection is the first imperative step of human face recognition systems. The two main problems of human face detection are how to cutoff the running time and how to reduce the number of false positives. In this paper, we present frontal and near-frontal face detection algorithm in still gray images using a thick edge image and neural network. We have devised a new filter that gets the thick edge image. Our overall scheme for face detection consists of two main phases. In the first phase we describe how to create the thick edge image using the filter and search for face candidates using a whole face detector. It is very helpful in removing plenty of windows with non-faces. The second phase verifies for detecting human faces using component-based eye detectors and the whole face detector. The experimental results show that our algorithm can reduce the running time and the number of false positives.

• 센서학회지
• /
• 제21권1호
• /
• pp.28-33
• /
• 2012

The visual sensing of human hands plays an important part in many man-machine interaction/interface systems. Most existing visionbased hand detection techniques depend on the color cues of human skin. The RGB color image from a vision sensor is often transformed to another color space as a preprocessing of hand detection because the color space transformation is assumed to increase the detection accuracy. However, the actual effect of color space transformation has not been well investigated in literature. This paper discusses a comparative evaluation of the pixel classification performance of hand skin detection in four widely used color spaces; RGB, YIQ, HSV, and normalized rgb. The experimental results indicate that using the normalized red-green color values is the most reliable under different backgrounds, lighting conditions, individuals, and hand postures. The nonlinear classification of pixel colors by the use of a multilayer neural network is also proposed to improve the detection accuracy.

• 한국정보전자통신기술학회논문지
• /
• 제4권3호
• /
• pp.202-208
• /
• 2011

최근 들어 인간과 컴퓨터의 상호작용을 통해 컴퓨터 시스템을 제어하는 기술에 관한 연구가 진행되고 있다. 이러한 응용분야의 대부분은 얼굴검출을 통해 사용자의 위치를 파악하고 사용자의 제스처를 인식하는 방법을 포함하고 있으나, 얼굴검출 성능은 아직 미흡한 실정이다. 사용자의 위치가 안정적으로 검출되지 못 하는 경우에는 제스처 인식 등의 인터페이스 성능은 현격하게 저하된다. 따라서 본 논문에서는 피부색과 얼굴검출의 누적 분포를 이용하여 동영상에서 안정적으로 얼굴을 검출할 수 있는 알고리즘을 제안하고, 실험을 통해 알고리즘의 유용성을 증명하였다. 제안한 알고리즘은 대응행렬 분석을 적용하여 사람을 추적하는 분야에 응용이 가능하다.

• 센서학회지
• /
• 제29권3호
• /
• pp.194-200
• /
• 2020

Human detection is an important aspect in many video-based sensing and monitoring systems. Studies have been actively conducted for the automatic detection of humans in camera images, and various methods have been proposed. However, there are still problems in terms of performance and computational cost. In this paper, we describe a method for efficient human detection in the field of view of a camera, which may be static or moving, through multiple processing steps. A detection line is designated at the position where a human appears first in a sensing area, and only the one-dimensional gray pixel values of the line are monitored. If any noticeable change occurs in the detection line, corner detection and optical flow computation are performed in the vicinity of the detection line to confirm the change. When significant changes are observed in the corner numbers and optical flow vectors, the final determination of human presence in the monitoring area is performed using the Histograms of Oriented Gradients method and a Support Vector Machine. The proposed method requires processing only specific small areas of two consecutive gray images. Furthermore, this method enables operation not only in a static condition with a fixed camera, but also in a dynamic condition such as an operation using a camera attached to a moving vehicle.

• 한국컴퓨터정보학회논문지
• /
• 제23권8호
• /
• pp.9-16
• /
• 2018

In this paper, a kidnapping detection scheme in which human pose estimation is used to classify accurately between kidnapping cases and normal ones is proposed. To estimate human poses from input video, human's 10 joint information is extracted by OpenPose library. In addition to the features which are used in the previous study to represent the size change rates and the regularities of human activities, the human pose estimation features which are computed from the location of detected human's joints are used as the features to distinguish kidnapping situations from the normal accompanying ones. A frame-based kidnapping detection scheme is generated according to the selection of J48 decision tree model from the comparison of several representative classification models. When a video has more frames of kidnapping situation than the threshold ratio after two people meet in the video, the proposed scheme detects and notifies the occurrence of kidnapping event. To check the feasibility of the proposed scheme, the detection accuracy of our newly proposed scheme is compared with that of the previous scheme. According to the experiment results, the proposed scheme could detect kidnapping situations more 4.73% correctly than the previous scheme.

• 센서학회지
• /
• 제25권4호
• /
• pp.285-290
• /
• 2016

Human sensing is an important but challenging technology. Unlike other methods for sensing humans, a vision sensor has many advantages, and there has been active research in automatic human detection in camera images. The combination of Histogram of Oriented Gradients (HOG) and Support Vector Machine (SVM) is currently one of the most successful methods in vision-based human detection. However, extracting HOG features from an image is computer intensive, and it is thus hard to employ the HOG method in real-time processing applications. This paper describes an efficient solution to this speed problem of the HOG method. Our method obtains edge information of an image and finds candidate regions where humans very likely exist based on the distribution pattern of the detected edge points. The HOG features are then extracted only from the candidate image regions. Since complex HOG processing is adaptively done by the guidance of the simpler edge detection step, human detection can be performed quickly. Experimental results show that the proposed method is effective in various images.

• 드라이브 ㆍ 컨트롤
• /
• 제17권3호
• /
• pp.1-8
• /
• 2020

Recently, a special purpose machine with two manipulators and quadruped crawler system has been developed for rapid life-saving and initial restoration work at disaster sites. This special purpose machine provides the driver with various environmental recognition functions for accurate and rapid task determination. In particular, the human detection technology assists the driver in poor working conditions such as low-light, dust, water vapor, fog, rain, etc. to prevent secondary human accidents when moving and working. In this study, a human detection module is developed to be mounted on a special purpose machine. A thermal sensor and CCD camera were used to detect victims and nearby workers in response to the difficult environmental conditions present at disaster sites. The performance of various AI-based life detection algorithm were verified and then applied to the task of detecting various objects with different postures and exposure conditions. In addition, image visibility improvement technology was applied to further improve the accuracy of human detection.

• 제어로봇시스템학회논문지
• /
• 제18권4호
• /
• pp.353-358
• /
• 2012

Human detection is a challenging task in many fields because it is difficult to detect humans due to their variable appearance and posture. Furthermore, it is also hard to track the detected human because of their dynamic and unpredictable behavior. The evaluation speed of method is also important as well as its accuracy. In this paper, we propose unified detection and tracking method for humans using Gaussian-PSO (Gaussian Particle Swarm Optimization) with the HOG (Histograms of Oriented Gradients) features to achieve a fast and accurate performance. Keeping the robustness of HOG features on human detection, we raise the process speed in detection and tracking so that it can be used for real-time applications. These advantages are given by a simple process which needs just one linear-SVM classifier with HOG features and Gaussian-PSO procedure for the both of detection and tracking.

• Journal of information and communication convergence engineering
• /
• 제21권1호
• /
• pp.98-102
• /
• 2023

This paper presents a study on how augmenting semi-synthetic image data improves the performance of human detection algorithms. In the field of object detection, securing a high-quality data set plays the most important role in training deep learning algorithms. Recently, the acquisition of real image data has become time consuming and expensive; therefore, research using synthesized data has been conducted. Synthetic data haves the advantage of being able to generate a vast amount of data and accurately label it. However, the utility of synthetic data in human detection has not yet been demonstrated. Therefore, we use You Only Look Once (YOLO), the object detection algorithm most commonly used, to experimentally analyze the effect of synthetic data augmentation on human detection performance. As a result of training YOLO using the Penn-Fudan dataset, it was shown that the YOLO network model trained on a dataset augmented with synthetic data provided high-performance results in terms of the Precision-Recall Curve and F1-Confidence Curve.

• 한국멀티미디어학회논문지
• /
• 제19권8호
• /
• pp.1345-1360
• /
• 2016

In this paper, we propose a new real-time human detection under omni-directional cameras for visual surveillance purpose, based on CNN with unified detection and AGMM. Compared to CNN-based state-of-the-art object detection methods. YOLO model-based object detection method boasts of very fast object detection, but with less accuracy. The proposed method adapts the unified detecting CNN of YOLO model so as to be intensified by the additional foreground contextual information obtained from pre-stage AGMM. Increased computational time incurred by additional AGMM processing is compensated by speed-up gain obtained from utilizing 2-D input data consisting of grey-level image data and foreground context information instead of 3-D color input data. Through various experiments, it is shown that the proposed method performs better with respect to accuracy and more robust to environment changes than YOLO model-based human detection method, but with the similar processing speeds to that of YOLO model-based one. Thus, it can be successfully employed for embedded surveillance application.