• Korean Journal of Remote Sensing
• /
• v.22 no.6
• /
• pp.543-551
• /
• 2006

Recently efforts to extract information about forests by using remote sensing techniques for efficient forest management have progressed actively. In terms of extraction of tree information using single remote sensing data, however, the accuracy of tree recognition and the quantity of extracted information is limited. The objective of this study is to carry out tree modeling in domestic environment applying the latest core technique for tree modeling using color aerial photographs and LiDAR data and to estimate the result of tree modeling. A small-scale coniferous forest was investigated in Daejeon. It was 0.77 that the $R^2$ of accuracy test of tree numbers that estimated with color aerial photography and LiDAR data. In terms of tree height, there was no difference between the estimated value and the field measurements in the case of the group accuracy test of the recently unchanged area. Moreover $R^2$ was 0.83 in the case of the individual accuracy test.

• Journal of the Korea Convergence Society
• /
• v.10 no.1
• /
• pp.103-113
• /
• 2019

License plate recognition camera is dedicated device designed for acquiring images of the target vehicle for recognizing letters and numbers in a license plate. Mostly, it is used as a part of the system combined with server and image analysis module rather than as a single use. However, building a system for vehicle license plate recognition is costly because it is required to construct a facility with a server providing the management and analysis of the captured images and an image analysis module providing the extraction of numbers and characters and recognition of the vehicle's plate. In this study, we would like to develop an embedded type convergent camera (Edge Base) which can expand the function of the camera to not only the license plate recognition but also the security CCTV function together and to perform two functions within the camera. This embedded type convergence camera equipped with a high resolution 4K IP camera for clear image acquisition and fast data transmission extracted license plate area by applying YOLO, a deep learning software for multi object recognition based on open source neural network algorithm and detected number and characters of the plate and verified the detection accuracy and recognition accuracy and confirmed that this camera can perform CCTV security function and vehicle number plate recognition function successfully.

• Korean Journal of Remote Sensing
• /
• v.34 no.6_1
• /
• pp.941-951
• /
• 2018

Utilization and demand of UAV (unmanned aerial vehicle) for the generation of 3D city model are increasing. In this study, we performed an experiment to adjustment position/orientation of UAV with incomplete attitude information and to extract point cloud data. In order to correct the attitude of the UAV, the rotation angle was calculated by using the continuous position information of UAV movements. Based on this, the corrected position/orientation information was obtained by applying IBA (Incremental Bundle Adjustment) based on photogrammetry. Each pair was transformed into an epipolar image, and the MDR (Multi-Dimensional Relaxation) technique was applied to obtain high precision DSM. Each extracted pair is aggregated and output in the form of a single point cloud or DSM. Using the DJI inspire1 and Phantom4 images, we can confirm that the point cloud can be extracted which expresses the railing of the building clearly. In the future, research will be conducted on improving the matching performance and establishing sensor models of oblique images. After that, we will continue the image processing technology for the generation of the 3D city model through the study of the extraction of 3D cloud It should be developed.

• Convergence Security Journal
• /
• v.19 no.2
• /
• pp.91-104
• /
• 2019

In this paper, we present a novel local descriptor, Local Prominent Directional Pattern (LPDP), to represent the description of facial images for gender recognition purpose. To achieve a clearly discriminative representation of local shape, presented method encodes a target pixel with the prominent directional variations in local structure from an analysis of statistics encompassed in the histogram of such directional variations. Use of the statistical information comes from the observation that a local neighboring region, having an edge going through it, demonstrate similar gradient directions, and hence, the prominent accumulations, accumulated from such gradient directions provide a solid base to represent the shape of that local structure. Unlike the sole use of gradient direction of a target pixel in existing methods, our coding scheme selects prominent edge directions accumulated from more samples (e.g., surrounding neighboring pixels), which, in turn, minimizes the effect of noise by suppressing the noisy accumulations of single or fewer samples. In this way, the presented encoding strategy provides the more discriminative shape of local structures while ensuring robustness to subtle changes such as local noise. We conduct extensive experiments on gender recognition datasets containing a wide range of challenges such as illumination, expression, age, and pose variations as well as sketch images, and observe the better performance of LPDP descriptor against existing local descriptors.

• Journal of odor and indoor environment
• /
• v.17 no.4
• /
• pp.355-361
• /
• 2018

Mold grows more easily when humidity is higher in indoor spaces, and as such is found more often on wetted areas in housing such as walls, toilets, kitchens, and poorly managed spaces. However, there have been few studies that have specifically assessed the level of mold in the indoor spaces of water-damaged housing in the Republic of Korea. We investigated the levels of airborne mold according to the characteristics of water damage types and explored the correlation between the distribution of mold genera and the characteristics of households. Samplings were performed from January 2016 to June 2018 in 97 housing units with water leakage or condensation, or a history of flooding, and in 61 general housing units in the metropolitan and Busan area, respectively. Airborne mold was collected on MEA (Malt extract agar) at flow rate of 100 L/min for 1 min. After collection, the samples were incubated at $25^{\circ}C$ for 120 hours. The cultured samples were counted and corrected using a positive hole conversion table. The samples were then analyzed by single colony culture, DNA extraction, gene amplification, and sequencing. By type of housing, concentrations of airborne mold were highest in flooded housing, followed by water-leaked or highly condensed housings, and then general housing. In more than 50% of water-damaged housing, the level of airborne mold exceeded the guideline of Korea's Ministry of Environment ($500CFU/m^3$). Of particular concern was the fact that the I/O ratio of water-damaged housing was greater than 1, which could indicate that mold damage may occur indoors. The distribution patterns of the fungal species were as follows: Penicillium spp., Cladosporium spp. (14%), Aspergillus spp. (13%) and Alternaria spp. (3%), but significant differences of their levels in indoor spaces were not found. Our findings indicate that high levels of mold damage were found in housing with water damage, and Aspergillus flavus and Penicillium brevicompactum were more dominant in housing with high water activity. Comprehensive management of flooded or water-damaged housing is necessary to reduce fungal exposure.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.2
• /
• pp.127-133
• /
• 2021

A recent study of people physically falling focused on analyzing the motions of the falls using a recurrent neural network (RNN) and a deep learning approach to get good results from detecting 2D human poses from a single color image. In this paper, we investigate a detection method for estimating the position of the head and shoulder keypoints and the acceleration of positional change using the skeletal keypoints information extracted using PoseNet from an image obtained with a low-cost 2D RGB camera, increasing the accuracy of judgments about the falls. In particular, we propose a fall detection method based on the characteristics of post-fall posture in the fall motion-analysis method. A public data set was used to extract human skeletal features, and as a result of an experiment to find a feature extraction method that can achieve high classification accuracy, the proposed method showed a 99.8% success rate in detecting falls more effectively than a conventional, primitive skeletal data-use method.

• Smart Media Journal
• /
• v.9 no.4
• /
• pp.66-72
• /
• 2020

As the online education market expands, educational contents with various presentation methods are being developed and released. In addition, it is imperative to develop content that reflects the usability and user environment of users who use this educational content. However, for qualitative growth of contents that will support quantitative expansion of markets, existing model analysis methods are urgently needed at a time when development direction of newly developed contents is secured. In this process of content development, a typical model for setting development goals is needed, as the rules of the prototype affect the entire development process and the final development outcome. It can also provide a positive benefit that screens the issue of performance dualization between processes due to the absence of communication between a single entity or between a number of entities. In the case of AR-based educational content which is effective to secure data necessary for development by securing samples of similar categories because there are not enough ready-made samples released. Therefore, a big data statistical analysis service is needed that can easily collect data and make decisions using big data. In this paper, we would like to design analysis services that enable the selection and detection of intuitive multidimensional factors and attributes, and propose big data-based statistical analysis services that can assist cooperative activities within an organization or among many companies.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.47 no.4
• /
• pp.317-331
• /
• 2021

Essential oil is a volatile substance obtained by physically obtaining fragrant plant materials made by one single plant and plant species, and is widely used for cosmetics, fragrances, and aroma therapy due to its excellent preservation, sterilization, and antibacterial effects. When essential oil would undergo the extraction and concentration processes, the agricultural chemicals thereof would be extracted and concentrated only to be harmful to the human body. This study analyzes 320 residual agricultural chemicals concentrated in the essential oil, and to this end, LC-MS/MS and GC-MS/MS are used, while the freezing process is applied instead of the conventional refining process hexane, to improve the preprocessing method. As a result of analyzing the essential oil, such ingredients as chlorpyrifos, piperonyl butoxide and silafluofen have been detected in Basil oil and Clove leaf oil. Hence, it is perceived that the residual agricultural chemicals should continue to be monitored for the essential oil.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.1-16
• /
• 2022

Despite recent breakthroughs in deep learning and computer vision fields, the pixel-wise identification of tiny objects in high-resolution images with complex disturbances remains challenging. This study proposes a modified U-net for tiny crack segmentation in real-world steel-box-girder bridges. The modified U-net adopts the common U-net framework and a novel Self-Attention-Self-Adaption (SASA) neuron as the fundamental computing element. The Self-Attention module applies softmax and gate operations to obtain the attention vector. It enables the neuron to focus on the most significant receptive fields when processing large-scale feature maps. The Self-Adaption module consists of a multiplayer perceptron subnet and achieves deeper feature extraction inside a single neuron. For data augmentation, a grid-based crack random elastic deformation (CRED) algorithm is designed to enrich the diversities and irregular shapes of distributed cracks. Grid-based uniform control nodes are first set on both input images and binary labels, random offsets are then employed on these control nodes, and bilinear interpolation is performed for the rest pixels. The proposed SASA neuron and CRED algorithm are simultaneously deployed to train the modified U-net. 200 raw images with a high resolution of 4928 × 3264 are collected, 160 for training and the rest 40 for the test. 512 × 512 patches are generated from the original images by a sliding window with an overlap of 256 as inputs. Results show that the average IoU between the recognized and ground-truth cracks reaches 0.409, which is 29.8% higher than the regular U-net. A five-fold cross-validation study is performed to verify that the proposed method is robust to different training and test images. Ablation experiments further demonstrate the effectiveness of the proposed SASA neuron and CRED algorithm. Promotions of the average IoU individually utilizing the SASA and CRED module add up to the final promotion of the full model, indicating that the SASA and CRED modules contribute to the different stages of model and data in the training process.

• Journal of Practical Engineering Education
• /
• v.14 no.3
• /
• pp.593-600
• /
• 2022

Various sensors such as lidar, radar, and camera are fused and used in autonomous vehicles. Rider and radar sensors are difficult to popularize because they are expensive equipment. In order to popularize autonomous vehicles, research that can replace expensive equipment is continuously being conducted. In this paper, we use a single camera that is inexpensive and can be easily mounted. We propose a method for detecting the wheels and adjacent lanes of a front-side vehicle of a driving vehicle and estimating distances. Our proposed method detects lanes and wheels from frame images after frame extraction via input images. In addition, the distance is measured and compared with the actual distance measured in the actual road environment. The distance could be calculated relatively accurately within the error range of ± 3 cm. Through this, it is expected that the camera can be used as an alternative means when the cost of autonomous vehicles is reduced or when the lidar or radar sensor fails.