• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.9
• /
• pp.2153-2160
• /
• 2015

An analysis of hardware design conditions of moving object detection (MOD) algorithm is described, which is based on effective Gaussian mixture learning (EGML). A simulation model of EGML algorithm is implemented using OpenCV, and the effects of some parameter values on background learning time and MOD sensitivity are analyzed for various images. In addition, optimal design conditions for hardware implementation of EGML-based MOD algorithm are extracted from fixed-point simulations for various bit-widths of parameters. The proposed fixed-point model of the EGML-based MOD uses only half of the bit-width at the expense of the loss of MOD performance within 0.5% when compared with floating-point MOD results.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2000.05a
• /
• pp.352-355
• /
• 2000

In this paper, we propose an adaptive thresholding for edge detection. first, we got histograms for background image and image with moving object, respectively. Then we make difference histogram between histograms of background and object image. A thresholding value is decided using gradient of peak to peak in the difference histogram. The experimentation is processed using a moving car in the road. The result is that edge is detected well regardless of the brightness.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.1039-1040
• /
• 2008

Multi-resolution approach to object detection wherein all entities including the partial information and background knowledge are modeled in set-theoretic terms whereby associated processing are formulated via set-theoretic operations is investigated. The generic set-theoretic paradigm is then applied to particular problems of detecting malfunctions in semiconductor fabrication process wherein the computational- and storage- efficiencies as enabled by morphological signal processing further coupled with flexibilities enabled by multi-resolution approach leads to a scalable paradigm in which the desired performance can be obtained on-demand fashion.

• Journal of Korea Multimedia Society
• /
• v.12 no.6
• /
• pp.794-803
• /
• 2009

In this paper, we will present a container auto-landing system, the system use the stereo camera to measure the container depth information. And the container region can be detected by using its hough line feature. In the line feature detection algorithm, we will detect the parallel lines and perpendicular lines which compose the rectangle region. Among all the candidate regions, we can select the region with the same aspect-ratio to the container. The region will be the detected container region. After having the object on both left and right images, we can estimate the distance from camera to object and container dimension. Then all the detect dimension information and depth inform will be applied to reconstruct the virtual environment of crane which will be introduce in this paper. Through the simulation result, we can know that, the container detection rate achieve to 97% with simple background. And the estimation algorithm can get a more accuracy result with a far distance than the near distance.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.64 no.3
• /
• pp.193-198
• /
• 2015

A large number of vision applications rely on matching keypoints across images. Pedestrian detection is under constant pressure to increase both its quality and speed. Such progress allows for new application. A higher speed enables its inclusion into large systems with extensive subsequent processing, and its deployment in computationally constrained scenarios. In this paper, we focus on improving the speed of pedestrian detection using HOG(histogram of oriented gradient) and multi frame operation which is robust to illumination changes in cluttering images. The result of our simulation indicates that the detection rate and speed of the proposed method is much faster than that of conventional HOG and differential images.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.26 no.2
• /
• pp.129-152
• /
• 2024

In the application of deep learning object detection via CCTV in tunnels, a large number of false positive detections occur due to the poor environmental conditions of tunnels, such as low illumination and severe perspective effect. This problem directly impacts the reliability of the tunnel CCTV-based accident detection system reliant on object detection performance. Hence, it is necessary to reduce the number of false positive detections while also enhancing the number of true positive detections. Based on a deep learning object detection model, this paper proposes a false positive data training method that not only reduces false positives but also improves true positive detection performance through retraining of false positive data. This paper's false positive data training method is based on the following steps: initial training of a training dataset - inference of a validation dataset - correction of false positive data and dataset composition - addition to the training dataset and retraining. In this paper, experiments were conducted to verify the performance of this method. First, the optimal hyperparameters of the deep learning object detection model to be applied in this experiment were determined through previous experiments. Then, in this experiment, training image format was determined, and experiments were conducted sequentially to check the long-term performance improvement through retraining of repeated false detection datasets. As a result, in the first experiment, it was found that the inclusion of the background in the inferred image was more advantageous for object detection performance than the removal of the background excluding the object. In the second experiment, it was found that retraining by accumulating false positives from each level of retraining was more advantageous than retraining independently for each level of retraining in terms of continuous improvement of object detection performance. After retraining the false positive data with the method determined in the two experiments, the car object class showed excellent inference performance with an AP value of 0.95 or higher after the first retraining, and by the fifth retraining, the inference performance was improved by about 1.06 times compared to the initial inference. And the person object class continued to improve its inference performance as retraining progressed, and by the 18th retraining, it showed that it could self-improve its inference performance by more than 2.3 times compared to the initial inference.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.13 no.4
• /
• pp.315-321
• /
• 2009

In this paper, a level set based energy functional is proposed, the minimization of which results in simultaneous reference background image modeling and foreground segmentation. Due to the mutual constraint of the two processes, a good estimate of the background can be obtained with a small number of frames, and due to the use of the level set, an Euler-Lagrange equation that directly solves the problem can be derived.

• Journal of Broadcast Engineering
• /
• v.12 no.3
• /
• pp.289-292
• /
• 2007

A background initialization algorithm based on the spatiotemporal similarity measure in a motion tracking system is proposed. From the accumulated difference between the base frame and the other frames in a subinterval, the regions affected by moving objects are located. The median is applied over the subsequence in the subinterval in which co-located regions share the similarity. The outputs from each subinterval are filtered by second stage median filter. The proposed method showed good results even in the busy and crowded sequences where the real background does not exit.

• Journal of Digital Convergence
• /
• v.10 no.6
• /
• pp.297-302
• /
• 2012

In this paper, we present a object detection and tracking method based on image context analysis. It is robust from the image variations such as complicated background, dynamic movement of the object. Image context analysis is carried out using the hybrid network of k-means and RBF. The proposed object detection employs context-driven adaptive Bayesian framework to relive the effect due to uneven object images. The proposed method used feature vector generator using 2D Haar wavelet transform and the Bayesian discriminant method in order to enhance the speed of learning. The system took less time to learn, and learning in a wide variety of data showed consistent results. After we developed the proposed method was applied to real-world environment. As a result, in the case of the object to detect pass outside expected area or other changes in the uncertain reaction showed that stable. The experimental results show that the proposed approach can achieve superior performance using various data sets to previously methods.

• Journal of Broadcast Engineering
• /
• v.16 no.3
• /
• pp.462-470
• /
• 2011

We proposed a realtime object tracking system that detects the abandoned or disappeared objects. Because these events are caused by human, we used the tracking based algorithm. After the background subtraction by Gaussian mixture model, the shadow removal is applied for accurate object detection. The static object is classified as either of abandoned objects or disappeared object. We assigned monitoring time to the static object to overcome a situation that it is being overlapped by other object. We obtained more accurate detection by using region growing method. We implemented our algorithm by DSP processor and obtained an excellent result throughout the experiment.