• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.9
• /
• pp.379-390
• /
• 2019

In this paper, we propose a novel deep neural network model for Vision-and-Language Navigation (VLN) named LVLN (Landmark-based VLN). In addition to both visual features extracted from input images and linguistic features extracted from the natural language instructions, this model makes use of information about places and landmark objects detected from images. The model also applies a context-based attention mechanism in order to associate each entity mentioned in the instruction, the corresponding region of interest (ROI) in the image, and the corresponding place and landmark object detected from the image with each other. Moreover, in order to improve the success rate of arriving the target goal, the model adopts a progress monitor module for checking substantial approach to the target goal. Conducting experiments with the Matterport3D simulator and the Room-to-Room (R2R) benchmark dataset, we demonstrate high performance of the proposed model.

• Journal of the Institute of Convergence Signal Processing
• /
• v.19 no.4
• /
• pp.192-198
• /
• 2018

Recently, the use of drone has been increasing rapidly in many ways. A drone can capture remote objects efficiently so it is suitable for surveillance and security systems. This paper discusses three methods for detecting moving vehicles using a drone. We compare three target detection methods using a background frame, preceding frames, or moving average frames. They are subtracted from a current frame. After the frame subtraction, morphological filters are applied to increase the detection rate and reduce the false alarm rate. In addition, the false alarm region is removed based on the true size of targets. In the experiments, three moving vehicles were captured by a drone, and the detection rate and the false alarm rate were obtained by three different methods and the results are compared.

• The Korean Journal of Pain
• /
• v.34 no.3
• /
• pp.339-345
• /
• 2021

Background: Ultrasound-guided caudal epidural injection (CEI) is limited in that it cannot confirm drug distribution at the target site without fluoroscopy. We hypothesized that visualization of solution flow through the inter-laminar space of the lumbosacral spine using color Doppler ultrasound alone would allow for confirmation of drug distribution. Therefore, we aimed to prospectively evaluate the usefulness of this method by comparing the color Doppler image in the paramedian sagittal oblique view of the lumbosacral spine (LS-PSOV) with the distribution of the contrast medium observed during fluoroscopy. Methods: Sixty-five patients received a 10-mL CEI of solution containing contrast medium under ultrasound guidance. During injection, flow was observed in the LSPSOV using color Doppler ultrasonography, following which it was confirmed using fluoroscopy. The presence of contrast image at L5-S1 on fluoroscopy was defined as "successful CEI." We then calculated prediction accuracy for successful CEI using color Doppler ultrasonography in the LS-PSOV. We also investigated the correlation between the distribution levels measured via color Doppler and fluoroscopy. Results: Prediction accuracy with color Doppler ultrasonography was 96.9%. The sensitivity, specificity, positive predictive value, and negative predictive value were 96.7%, 100%, 100%, and 60.0%, respectively. In 52 of 65 patients (80%), the highest level at which contrast image was observed was the same for both color Doppler ultrasonography and fluoroscopy. Conclusions: Our findings demonstrate that color Doppler ultrasonography in the LS-PSOV is a new method for determining whether a drug solution reaches the lumbosacral region (i.e., the main target level) without the need for fluoroscopy.

• Journal of Biomedical Engineering Research
• /
• v.43 no.3
• /
• pp.161-169
• /
• 2022

Engagement is an important factor in the field of rehabilitation as it is a known factor that have a positive influence on functional gaining in people who receive rehabilitation therapy. Although a number of measurements for engagement have been recently developed, investigation of possible factors that may have influence on engagement is not well established. Currently available evidence suggests that engagement is affected by mood and it is hypothesized that a personal factor may contribute to engagement. Therefore, this study aims to test the hypothetical relationship between mood and engagement while performing a manual dexterity task through an experiment in healthy participants prior to investigation on people with medical condition who requires rehabilitation therapy. After inducing target mood (positive or negative mood) for study participants by asking them to recall autobiographical memories, change in muscle activity, which was operationalized as an indicator of engagement, was investigated. Electromyogram (EMG) was recorded from four muscle areas in non-dominant hand side to quantify muscle activity. The results show that the target moods were appropriately induced with the method. Although there were subtle differences in the level of engagement between different moods, certain variables derived from muscle activity were significantly different; mean amplitude for wrist extensor EMG showed significant difference between different moods (Z = -2.023, p < .05) indicating that muscle activities in the wrist extensor are greater for positive mood than negative mood region during manual dexterity task. Meanwhile, performance outcomes of Minnesota Manual Dexterity Test (MMDT), such as mean completion time and number of errors, between moods showed no significant difference in two different moods, resulting in MMDT administration may not be useful task in distinguishing the level of rehabilitation engagement.

• 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.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.421-436
• /
• 2023

This paper presents a vibration displacement measurement and damage identification method for a space truss structure from its vibration videos. Features from Accelerated Segment Test (FAST) algorithm is combined with adaptive threshold strategy to detect the feature points of high quality within the Region of Interest (ROI), around each node of the truss structure. Then these points are tracked by Kanade-Lucas-Tomasi (KLT) algorithm along the video frame sequences to obtain the vibration displacement time histories. For some cases with the image plane not parallel to the truss structural plane, the scale factors cannot be applied directly. Therefore, these videos are processed with homography transformation. After scale factor adaptation, tracking results are expressed in physical units and compared with ground truth data. The main operational frequencies and the corresponding mode shapes are identified by using Subspace Stochastic Identification (SSI) from the obtained vibration displacement responses and compared with ground truth data. Structural damages are quantified by elemental stiffness reductions. A Bayesian inference-based objective function is constructed based on natural frequencies to identify the damage by model updating. The Success-History based Adaptive Differential Evolution with Linear Population Size Reduction (L-SHADE) is applied to minimise the objective function by tuning the damage parameter of each element. The locations and severities of damage in each case are then identified. The accuracy and effectiveness are verified by comparison of the identified results with the ground truth data.

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.1
• /
• pp.1-11
• /
• 2023

Deep-learning-based image segmentation is one of the most widely employed lane detection approaches, and it requires a post-process for extracting the key points on the lanes. A general approach for key-point extraction is using a fixed threshold defined by a user. However, finding the best threshold is a manual process requiring much effort, and the best one can differ depending on the target data set (or an image). We propose a novel key-point extraction algorithm that automatically adapts to the target image without any manual threshold setting. In our adaptive key-point extraction algorithm, we propose a line-level normalization method to distinguish the lane region from the background clearly. Then, we extract a representative key point for each lane at a line (row of an image) using a kernel density estimation. To check the benefits of our approach, we applied our method to two lane-detection data sets, including TuSimple and CULane. As a result, our method achieved up to 1.80%p and 17.27% better results than using a fixed threshold in the perspectives of accuracy and distance error between the ground truth key-point and the predicted point.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.877-885
• /
• 2022

Vision-based activity recognition has been widely attempted at construction sites to estimate productivity and enhance workers' health and safety. Previous studies have focused on extracting an individual worker's postural information from sequential image frames for activity recognition. However, various trades of workers perform different tasks with similar postural patterns, which degrades the performance of activity recognition based on postural information. To this end, this research exploited a concept of human-object interaction, the interaction between a worker and their surrounding objects, considering the fact that trade workers interact with a specific object (e.g., working tools or construction materials) relevant to their trades. This research developed an approach to understand the context from sequential image frames based on four features: posture, object, spatial features, and temporal feature. Both posture and object features were used to analyze the interaction between the worker and the target object, and the other two features were used to detect movements from the entire region of image frames in both temporal and spatial domains. The developed approach used convolutional neural networks (CNN) for feature extractors and activity classifiers and long short-term memory (LSTM) was also used as an activity classifier. The developed approach provided an average accuracy of 85.96% for classifying 12 target construction tasks performed by two trades of workers, which was higher than two benchmark models. This experimental result indicated that integrating a concept of the human-object interaction offers great benefits in activity recognition when various trade workers coexist in a scene.

• Clinical Pain
• /
• v.18 no.2
• /
• pp.59-64
• /
• 2019

Objective: This study evaluated the feasibility of ultrasound-guided lumbar nerve root block (LNRB) and S1 nerve root block by identifying spread patterns via fluoroscopy in cadavers. Method: A total of 48 ultrasound-guided injections were performed in 4 fresh cadavers from L1 to S1 roots. The target point of LNRB was the midpoint between the lower border of the transverse process and the facet joint at each level. The target point of S1 nerve root block was the S1 foramen, which can be visualized between the median sacral crest and the posterior superior iliac spine, below the L5-S1 facet joint. The injection was performed via an in-plane approach under real-time axial view ultrasound guidance. Fluoroscopic validation was performed after the injection of 2 cc of contrast agent. Results: The needle placements were correct in all injections. Fluoroscopy confirmed an intra-foraminal contrast spreading pattern following 41 of the 48 injections (85.4%). The other 7 injections (14.6%) yielded typical neurograms, but also resulted in extra-foraminal patterns that occurred evenly in each nerve root, including S1. Conclusion: Ultrasound-guided injection may be an option for the delivery of injectate into the S1 nerve root, as well as lumbar nerve root area.

• Genomics & Informatics
• /
• v.21 no.4
• /
• pp.52.1-52.10
• /
• 2023

Accurate and efficient microbial diagnosis is crucial for effective molecular diagnostics, especially in the field of human healthcare. The gold standard equipment widely employed for detecting specific microorganisms in molecular diagnosis is quantitative real-time polymerase chain reaction (qRT-PCR). However, its limitations in low metagenomic DNA yield samples necessitate exploring alternative approaches. Digital PCR, by quantifying the number of copies of the target sequence, provides absolute quantification results for the bacterial strain. In this study, we compared the diagnostic efficiency of qRT-PCR and digital PCR in detecting a particular bacterial strain (Staphylococcus aureus), focusing on skin-derived DNA samples. Experimentally, specific primer for S. aureus were designed at transcription elongation factor (greA) gene and the target amplicon were cloned and sequenced to validate efficiency of specificity to the greA gene of S. aureus. To quantify the absolute amount of microorganisms present on the skin, the variable region 5 (V5) of the 16S rRNA gene was used, and primers for S. aureus identification were used to relative their amount in the subject's skin. The findings demonstrate the absolute convenience and efficiency of digital PCR in microbial diagnostics. We suggest that the high sensitivity and precise quantification provided by digital PCR could be a promising tool for detecting specific microorganisms, especially in skin-derived DNA samples with low metagenomic DNA yields, and that further research and implementation is needed to improve medical practice and diagnosis.