• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.6
• /
• pp.778-786
• /
• 2002

Scene boundary detection is important in the understanding of semantic structure from video data. However, it is more difficult than shot change detection because scene boundary detection needs to understand semantics in video data well. In this paper, we propose a new approach to scene segmentation using contextual information in video data. The contextual information is divided into two categories: local and global contextual information. The local contextual information refers to the foreground regions' information, background and shot activity. The global contextual information refers to the video shot's environment or its relationship with other video shots. Coherence, interaction and the tempo of video shots are computed as global contextual information. Using the proposed contextual information, we detect scene boundaries. Our proposed approach consists of three consecutive steps: linking, verification, and adjusting. We experimented the proposed approach using TV dramas and movies. The detection accuracy of correct scene boundaries is over than 80%.

• Medical Lasers
• /
• v.9 no.2
• /
• pp.159-165
• /
• 2020

Background and Objectives The ultimate goal in current skin rejuvenation practice is to achieve a good result with minimal pain and downtime. Nonablative skin rejuvenation (NSR) is one technique. The efficacy of the long-pulsed 1064 nm Nd:YAG laser (LPNDY) has not been assessed in NSR. Materials and Methods Three target areas were selected (bilateral cheeks and glabellar region) in six volunteer subjects. A LPNDY with an integral skin temperature monitor delivered three stacked shots to each target area (1064 nm, 12 mm spot, 13 J/cm², 1 Hz) without any skin cooling or anesthesia. The skin temperature was recorded before, during, and after each set of shots using the system monitor and in real-time using a high-sensitivity (±0.001℃) near-infrared video camera. The skin reaction was observed with the naked eye, and pain and discomfort were assessed by the subjects during and after treatment. Results The subjects reported a mild feeling of heat with no discomfort during or after the test treatments. Mild erythema was observed around the treatment areas, without noticeable edema. A series of three ascending skin temperature stepwise peaks, with a decrease in skin temperature towards the baseline after the third shot, was observed consistently. The mean temperatures for shots 1, 2, and 3 for the cheeks were 39.5℃, 42.0℃, and 44.4℃, respectively, and for the glabella, 40.8℃, 43.9℃, and 46.2℃, respectively. Similar ranges were indicated on the system integral temperature monitor. Conclusion A set of three stacked pulses with the LPNDY at a low fluence achieved ideal dermal temperatures to achieve some dermal remodeling but without any downtime or adverse events. The temperature data from the integral thermal sensor matched the video camera measurements with practical accuracy for skin rejuvenation requirements. These data suggest that LPNDY would satisfy the necessary criteria to achieve effective NSR, but further studies will be needed to assess the actual results in clinical practice.

• Journal of Korean Neurosurgical Society
• /
• v.52 no.3
• /
• pp.204-209
• /
• 2012

Objective : The authors performed a retrospective study to assess the accuracy and clinical benefits of a navigation coupled with O-arm$^{(R)}$ system guided method in the thoracic and lumbar spines by comparing with a C-arm fluoroscopy-guided method. Methods : Under the navigation guidance, 106 pedicle screws inserted from T7 to S1 in 24 patients, and using the fluoroscopy guidance, 204 pedicle screws from T5 to S1 in 45 patients. The position of screws within the pedicle was classified into four groups, from grade 0 (no violation cortex) to 3 (more than 4 mm violation). The location of violated pedicle cortex was also assessed. Intra-operative parameters including time required for preparation of screwing procedure, times for screwing and the number of X-ray shot were assessed in each group. Results : Grade 0 was observed in 186 (91.2%) screws of the fluoroscopy-guided group, and 99 (93.4%) of the navigation-guided group. Mean time required for inserting a screw was 3.8 minutes in the fluoroscopy-guided group, and 4.5 minutes in the navigation-guided group. Mean time required for preparation of screw placement was 4 minutes in the fluoroscopy-guided group, and 19 minutes in the navigation-guided group. The fluoroscopy-guided group required mean 8.9 times of X-ray shot for each screw placement. Conclusion : The screw placement under the navigation-guidance coupled with O-arm$^{(R)}$ system appears to be more accurate and safer than that under the fluoroscopy guidance, although the preparation and screwing time for the navigation-guided surgery is longer than that for the fluoroscopy-guided surgery.

• Journal of the Institute of Convergence Signal Processing
• /
• v.5 no.4
• /
• pp.294-300
• /
• 2004

According to the development of digital media technologies various algorithms for video clip matching have been proposed to match the video sequences efficiently. A large number of video search methods have focused on frame-wise query, whereas a relatively few algorithms have been presented for video clip matching or video shot matching. In this paper, we propose an efficient algorithm to index the video sequences and to retrieve the sequences for video clip query. To improve the accuracy and performance of video sequence matching, we employ the Cauchy function as a similarity measure between histograms of consecutive frames, which yields a high performance compared with conventional measures. The key frames extracted from segmented video shots can be used not only for video shot clustering but also for video sequence matching or browsing, where the key frame is defined by the frame that is significantly different from the previous frames. Experimental results with color video sequences show that the proposed method yields the high matching performance and accuracy with a low computational load compared with conventional algorithms.

• IEIE Transactions on Smart Processing and Computing
• /
• v.6 no.3
• /
• pp.193-199
• /
• 2017

Instead of using only spatial features from a single frame for person re-identification, a combination of spatial and temporal factors boosts the performance of the system. A recurrent neural network (RNN) shows its effectiveness in generating highly discriminative sequence-level human representations. In this work, we implement RNN, three Long Short Term Memory (LSTM) network variants, and Gated Recurrent Unit (GRU) on Caffe deep learning framework, and we then conduct experiments to compare performance in terms of size and accuracy for person re-identification. We propose using GRU for the optimized choice as the experimental results show that the GRU achieves the highest accuracy despite having fewer parameters than the others.

• Journal of the Korean Society for Precision Engineering
• /
• v.31 no.8
• /
• pp.671-675
• /
• 2014

Nozzle ring is an important part of turbocharger which is applied to today's most diesel engines. Turbo charger nozzle ring is difficult to process and takes a high cost and a long time relatively. For this reason, it is largely produced by using a precision casting. Investment method, the representative technology of precision casting, has excellent dimensional accuracy and can produce complex shapes relatively easily. However, it is difficult to avoid the casting defects such as shrinkage cavity and short shot. This study is to predict the casting defects which could be occurred during the investment method by use of finite element analysis software and to design the process and mold of the marine turbocharger nozzle ring.

• Smart Structures and Systems
• /
• v.23 no.4
• /
• pp.359-371
• /
• 2019

As part of a structural health monitoring system, the relative geometric relationship between a ship and bridge has been recognized as important for bridge authorities and ship owners to avoid ship-bridge collision. This study proposes a novel computer vision method for the real-time geometric parameter identification of moving ships based on a single shot multibox detector (SSD) by using transfer learning techniques and monocular vision. The identification framework consists of ship detection (coarse scale) and geometric parameter calculation (fine scale) modules. For the ship detection, the SSD, which is a deep learning algorithm, was employed and fine-tuned by ship image samples downloaded from the Internet to obtain the rectangle regions of interest in the coarse scale. Subsequently, for the geometric parameter calculation, an accurate ship contour is created using morphological operations within the saturation channel in hue, saturation, and value color space. Furthermore, a local coordinate system was constructed using projective geometry transformation to calculate the geometric parameters of ships, such as width, length, height, localization, and velocity. The application of the proposed method to in situ video images, obtained from cameras set on the girder of the Wuhan Yangtze River Bridge above the shipping channel, confirmed the efficiency, accuracy, and effectiveness of the proposed method.

• Journal of Broadcast Engineering
• /
• v.26 no.7
• /
• pp.844-854
• /
• 2021

Semi-supervised learning (SSL) and few-shot learning (FSL) have shown impressive performance even then the volume of labeled data is very limited. However, SSL and FSL can encounter a significant performance degradation if the diversity gap between the labeled and unlabeled data is high. To reduce this diversity gap, we propose a novel scheme that relies on an autoencoder for generating pseudo examples. Specifically, the autoencoder is trained on a specific class using the available labeled data and the decoder of the trained autoencoder is then used to generate N samples of that specific class based on N random noise, sampled from a standard normal distribution. The above process is repeated for all the classes. Consequently, the generated data reduces the diversity gap and enhances the model performance. Extensive experiments on MNIST and FashionMNIST datasets for SSL and FSL verify the effectiveness of the proposed approach in terms of classification accuracy and robustness against adversarial attacks.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.5
• /
• pp.45-52
• /
• 2004

According tothe development of digital media technologies various algorithms for video sequence matching have been proposed to match the video sequences efficiently. A large number of video sequence matching methods have focused on frame-wise query, whereas a relatively few algorithms have been presented for video sequence matching or video shot matching. In this paper, we propose an efficientalgorithm to index the video sequences and to retrieve the sequences for video sequence query. To improve the accuracy and performance of video sequence matching, we employ the Cauchy function as a similarity measure between histograms of consecutive frames, which yields a high performance compared with conventional measures. The key frames extracted from segmented video shots can be used not only for video shot clustering but also for video sequence matching or browsing, where the key frame is defined by the frame that is significantly different from the previous fames. Several key frame extraction algorithms have been proposed, in which similar methods used for shot boundary detection were employed with proper similarity measures. In this paper, we propose the efficient algorithm to extract key frames using the cumulative Cauchy function measure and. compare its performance with that of conventional algorithms. Video sequence matching can be performed by evaluating the similarity between data sets of key frames. To improve the matching efficiency with the set of extracted key frames we employ the Cauchy function and the modified Hausdorff distance. Experimental results with several color video sequences show that the proposed method yields the high matching performance and accuracy with a low computational load compared with conventional algorithms.

• International Journal of Internet, Broadcasting and Communication
• /
• v.11 no.4
• /
• pp.76-85
• /
• 2019

We introduce a vision-based object detection method for real-time video surveillance system in low-end edge computing environments. Recently, the accuracy of object detection has been improved due to the performance of approaches based on deep learning algorithm such as Region Convolutional Neural Network(R-CNN) which has two stage for inferencing. On the other hand, one stage detection algorithms such as single-shot detection (SSD) and you only look once (YOLO) have been developed at the expense of some accuracy and can be used for real-time systems. However, high-performance hardware such as General-Purpose computing on Graphics Processing Unit(GPGPU) is required to still achieve excellent object detection performance and speed. To address hardware requirement that is burdensome to low-end edge computing environments, We propose sub-frame analysis method for the object detection. In specific, We divide a whole image frame into smaller ones then inference them on Convolutional Neural Network (CNN) based image detection network, which is much faster than conventional network designed forfull frame image. We reduced its computationalrequirementsignificantly without losing throughput and object detection accuracy with the proposed method.