• Journal of Broadcast Engineering
• /
• v.25 no.6
• /
• pp.882-897
• /
• 2020

Recently, the use of immersive media contents representing Panorama and 360° video is increasing. Since the viewing angle is limited to generate the content through a general camera, image stitching is mainly used to combine images taken with multiple cameras into one image having a wide field of view. However, if the parallax between the cameras is large, parallax distortion may occur in the stitched image, which disturbs the user's content immersion, thus an image stitching overcoming parallax distortion is required. The existing Seam Optimization based image stitching method to overcome parallax distortion uses energy function or object segment information to reflect the location information of objects, but the initial seam generation location, background information, performance of the object detector, and placement of objects may limit application. Therefore, in this paper, we propose an image stitching method that can overcome the limitations of the existing method by adding a weight value set differently according to the type of object to the energy value using object detection based on deep learning.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.41 no.1
• /
• pp.25-32
• /
• 2004

Nowadays image processing is very useful for some field of traffic applications. The one reason is we can construct the system in a low price, the other is the improvement of hardware processing power, it can be more fast to processing the data. In traffic field, the development of image using system is interesting issue. Because it has the advantage of price of installation and it does not obstruct traffic during the installation. In this study, 1 propose the traffic monitoring system that implement on the embedded system environment. The whole system consists of two main part, one is host controller board, the other is image processing board. The part of host controller board take charge of control the total system interface of external environment, and OSD(On screen display). The part of image processing board takes charge of image input and output using video encoder and decoder, Image classification and memory control of using FPGA, control of mouse signal. And finally, for stable operation of host controller board, uC/OS-II operating system is ported on the board.

• Journal of Korean Academy of Oral and Maxillofacial Radiology
• /
• v.22 no.2
• /
• pp.273-282
• /
• 1992

The purpose of this study was to propose the utility which was evaluated the digital image processing and clinical application of the videodensitomery. The experiments were performed with IBM-PC/16bit-AT compatible, video camera(CCdtr55, Sony Co., Japan), an color monitor(MultiSync 3D, NEC, Japan) providing the resolution of 512×480 and 64 levels of gray. Sylvia Image Capture Board for the ADC(analog to digital converter) was used, composed of digitized image from digital signal and the radiographic density was measured by 256 level of gray. The periapical radiograph(Ektaspeed EP-21, Kodak Co., U. S. A) which was radiographed dried human mandible by exposure condition of 70 kVp and 48 impulses, was used for primary X-ray detector. And them evaluated for digitzed image by low and high pass filtering, correlations between aluminum equivalent values and the thickness of aluminum step wedge, aluminum equivalent values of sound enamel, dentin, and alveolar bone, the range of diffuse density for gray level ranging from 0 to 255. The obtained results were as follows: 1. The edge between aluminum steps of digitized image were somewhat blurred by low pass filtering, but edge enhancement could be resulted by high pass filtering. Expecially, edge enhancement between distal root of lower left 2nd molar and alveolar lamina dura was observed. 2. The correlation between aluminum equivalent values and the thickness of aluminum step wedge was intimated, yielding the coefficient of correlation r=0.9997(p<0.00l), the regression line was described by Y=0.9699X+0.456, and coefficient of variation amounting to 1.5%. 3. The aluminum equivalent values of sound enamel, dentin, and alvolar bone were 15.41㎜, 12.48㎜, 10.35㎜, respectively. 4. The range of diffuse density for gray level ranging from 0 to 255 was wider enough than that of photodenstiometer to be within the range of 1-4.9.

• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.478-481
• /
• 2005

The PMU (Payload Management Unit) is the main subsystem for the management, control and power supply of the MSC (Multi-Spectral Camera) Payload operation. It is the most important function for the electro-optical camera system that performs the Non-Uniformity Correction (NUC) function of the raw imagery data, rearranges the data from the CCD (Charge Coupled Device) detector and output it to the Data Compression and Storage Unit (DCSU). The NUC board in PMU performs it. In this paper, the NUC board system is described in terms of the configuration and the function, the efficiency for non-uniformity correction, and the influence of the data compression upon the peculiar feature of the CCD pixel. The NUC board is an image-processing unit within the PMU that receives video data from the CEV (Camera Electronic Unit) boards via a hotlinkand performs non-uniformity corrections upon the pixels according to commands received from the SBC (Single Board Computer) in the PMU. The lossy compression in DCSU needs the NUC in on-orbit condition.

• Journal of Korea Multimedia Society
• /
• v.13 no.7
• /
• pp.983-990
• /
• 2010

The image processing technique is very sensitive to the variation of external environment, so it tends to lose a lot of accuracy when the external environment changes rapidly. In this paper, we propose a vehicle detecting and tracking system for crime prevention suitable for an external environments with various changes using the image processing technique. Because the vehicle camera detector for crime prevention extracts and tracks the vehicle within one lane, it is important to classify a characteristic region rather than the contour of a vehicle. The proposed system detects the entrance of the vehicle using optical flow and tracks the vehicle by classifying the headlights, the bonnet, the front-window and the roof area of the vehicle. Experimental results show that the proposed method is robust to the environmental changes such as type, speed and time of a vehicle.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.1
• /
• pp.44-49
• /
• 2021

Recently, due to COVID-19, studies have been popularly worked to apply neural network to mask wearing automatic detection system. For applying neural networks, the 1-stage detection or 2-stage detection methods are used, and if data are not sufficiently collected, the pretrained neural network models are studied by applying fine-tuning techniques. In this paper, the system is consisted of 2-stage detection method that contain MTCNN model for face recognition and ResNet model for mask detection. The mask detector was experimented by applying five ResNet models to improve accuracy and fps in various environments. Training data used 17,217 images that collected using web crawler, and for inference, we used 1,913 images and two one-minute videos respectively. The experiment showed a high accuracy of 96.39% for images and 92.98% for video, and the speed of inference for video was 10.78fps.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.7
• /
• pp.41-51
• /
• 2024

In this paper, we explore the enhancement of target detection accuracy in the guided weapon using deep learning object detection on infrared (IR) images. Due to the characteristics of IR images being influenced by factors such as time and temperature, it's crucial to ensure a consistent representation of object features in various environments when training the model. A simple way to address this is by emphasizing the features of target objects and reducing noise within the infrared images through appropriate pre-processing techniques. However, in previous studies, there has not been sufficient discussion on pre-processing methods in learning deep learning models based on infrared images. In this paper, we aim to investigate the impact of image pre-processing techniques on infrared image-based training for object detection. To achieve this, we analyze the pre-processing results on infrared images that utilized global or local information from the video and the image. In addition, in order to confirm the impact of images converted by each pre-processing technique on object detector training, we learn the YOLOX target detector for images processed by various pre-processing methods and analyze them. In particular, the results of the experiments using the CLAHE (Contrast Limited Adaptive Histogram Equalization) shows the highest detection accuracy with a mean average precision (mAP) of 81.9%.

• Journal of Broadcast Engineering
• /
• v.24 no.3
• /
• pp.495-505
• /
• 2019

Typical algorithms of deep learning include CNN(Convolutional Neural Networks), which are mainly used for image recognition, and RNN(Recurrent Neural Networks), which are used mainly for speech recognition and natural language processing. Among them, CNN is able to learn from filters that generate feature maps with algorithms that automatically learn features from data, making it mainstream with excellent performance in image recognition. Since then, various algorithms such as R-CNN and others have appeared in object detection to improve performance of CNN, and algorithms such as YOLO(You Only Look Once) and SSD(Single Shot Multi-box Detector) have been proposed recently. However, since these deep learning-based detection algorithms determine the success of the detection in the still images, stable object tracking and detection in the video requires separate tracking capabilities. Therefore, this paper proposes a method of combining Kalman filters into deep learning-based detection networks for improved object tracking and detection performance in the video. The detection network used YOLO v2, which is capable of real-time processing, and the proposed method resulted in 7.7% IoU performance improvement over the existing YOLO v2 network and 20 fps processing speed in FHD images.

• Journal of Korea Multimedia Society
• /
• v.22 no.2
• /
• pp.136-145
• /
• 2019

The crowded environment of a domestic pig farm is highly vulnerable to the spread of infectious diseases such as foot-and-mouth disease, and studies have been conducted to automatically analyze behavior of pigs in a crowded pig farm through a video surveillance system using a camera. Although it is required to correctly separate occluding pigs for tracking each individual pigs, extracting the boundaries of the occluding pigs fast and accurately is a challenging issue due to the complicated occlusion patterns such as X shape and T shape. In this study, we propose a fast and accurate method to separate occluding pigs not only by exploiting the characteristics (i.e., one of the fast deep learning-based object detectors) of You Only Look Once, YOLO, but also by overcoming the limitation (i.e., the bounding box-based object detector) of YOLO with the test-time data augmentation of rotation. Experimental results with two-pigs occlusion patterns show that the proposed method can provide better accuracy and processing speed than one of the state-of-the-art widely used deep learning-based segmentation techniques such as Mask R-CNN (i.e., the performance improvement over Mask R-CNN was about 11 times, in terms of the accuracy/processing speed performance metrics).

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.3
• /
• pp.11-22
• /
• 2010

This paper is a study on implementation of intelligent image surveillance system using context information and supplements temporal-spatial constraint, the weak point in which it is hard to process it in real time. In this paper, we propose scene analysis algorithm which can be processed in real time in various environments at low resolution video(320*240) comprised of 30 frames per second. The proposed algorithm gets rid of background and meaningless frame among continuous frames. And, this paper uses wavelet transform and edge histogram to detect shot boundary. Next, representative key-frame in shot boundary is selected by key-frame selection parameter and edge histogram, mathematical morphology are used to detect only motion region. We define each four basic contexts in accordance with angles of feature points by applying vertical and horizontal ratio for the motion region of detected object. These are standing, laying, seating and walking. Finally, we carry out scene analysis by defining simple context model composed with general context and emergency context through estimating each context's connection status and configure a system in order to check real time processing possibility. The proposed system shows the performance of 92.5% in terms of recognition rate for a video of low resolution and processing speed is 0.74 second in average per frame, so that we can check real time processing is possible.