• Proceedings of the Korea Information Processing Society Conference
• /
• 한국정보처리학회 2023년도 춘계학술발표대회
• /
• pp.615-617
• /
• 2023

In recent years, the growing popularity of dogs due to the benefits they bring their owners has contributed to the increase of the number of dogs raised. For owners, it is their responsibility to ensure their dogs' health and safety. However, it is challenging for them to continuously monitor their dogs' activities, which are important to understand and guarantee their wellbeing. In this work, we introduce a camera-based monitoring system to help owners automatically monitor their dogs' activities. The system receives sequences of RGB images and uses YOLOv7 to detect the dog presence, and then applies post-processing to perform dog-centered image cropping on each input sequence. The optical flow is extracted from each sequence, and both sequences of RGB and flow are input to a two-stream EfficientNet to extract their respective features. Finally, the features are concatenated, and a bi-directional LSTM is utilized to retrieve temporal features and recognize the activity. The experiments prove that our system achieves a good performance with the F-1 score exceeding 0.90 for all activities and reaching 0.963 on average.

• Journal of Broadcast Engineering
• /
• 제23권5호
• /
• pp.598-605
• /
• 2018

Behavior awareness is a technology that recognizes human behavior through data and can be used in applications such as risk behavior through video surveillance systems. Conventional behavior recognition algorithms have been performed using the 2D camera image device or multi-mode sensor or multi-view or 3D equipment. When two-dimensional data was used, the recognition rate was low in the behavior recognition of the three-dimensional space, and other methods were difficult due to the complicated equipment configuration and the expensive additional equipment. In this paper, we propose a method of recognizing human behavior using only CCTV images without additional equipment using only RGB and depth information. First, the skeleton extraction algorithm is applied to extract points of joints and body parts. We apply the equations to transform the vector including the displacement vector and the relational vector, and study the continuous vector data through the RNN model. As a result of applying the learned model to various data sets and confirming the accuracy of the behavior recognition, the performance similar to that of the existing algorithm using the 3D information can be verified only by the 2D information.

• IEMEK Journal of Embedded Systems and Applications
• /
• 제7권4호
• /
• pp.163-172
• /
• 2012

A driver assistance system is critical to improve a convenience and stability of vehicle driving. Several systems have been already commercialized such as adaptive cruise control system and forward collision warning system. Efficient vehicle detection is very important to improve such driver assistance systems. Most existing vehicle detection systems are based on a radar system, which measures distance between a host and leading (or oncoming) vehicles under various weather conditions. However, it requires high deployment cost and complexity overload when there are many vehicles. A camera based vehicle detection technique is also good alternative method because of low cost and simple implementation. In general, night time vehicle detection is more complicated than day time vehicle detection, because it is much more difficult to distinguish the vehicle's features such as outline and color under the dim environment. This paper proposes a method to detect vehicles at night time using analysis of a captured color space with reduction of reflection and other light sources in images. Four colors spaces, namely RGB, YCbCr, normalized RGB and Ruta-RGB, are compared each other and evaluated. A suboptimal threshold value is determined by Otsu algorithm and applied to extract candidates of taillights of leading vehicles. Statistical features such as mean, variance, skewness, kurtosis, and entropy are extracted from the candidate regions and used as feature vector for SVM(Support Vector Machine) classifier. According to our simulation results, the proposed statistical feature based SVM provides relatively high performances of leading vehicle detection with various distances in variable nighttime environments.

• Journal of The Korean Society of Agricultural Engineers
• /
• 제66권4호
• /
• pp.27-39
• /
• 2024

Soil texture is determined by the proportions of sand, silt, and clay within the soil, which influence characteristics such as porosity, water retention capacity, electrical conductivity (EC), and pH. Traditional classification of soil texture requires significant sample preparation including oven drying to remove organic matter and moisture, a process that is both time-consuming and costly. This study aims to explore an alternative method by developing an AI model capable of predicting soil texture from images of pre-sorted soil samples using computer vision and deep learning technologies. Soil samples collected from agricultural fields were pre-processed using sieve analysis and the images of each sample were acquired in a controlled studio environment using a smartphone camera. Color distribution ratios based on RGB values of the images were analyzed using the OpenCV library in Python. A convolutional neural network (CNN) model, built on PyTorch, was enhanced using Digital Image Processing (DIP) techniques and then trained across nine distinct conditions to evaluate its robustness and accuracy. The model has achieved an accuracy of over 80% in classifying the images of pre-sorted soil samples, as validated by the components of the confusion matrix and measurements of the F1 score, demonstrating its potential to replace traditional experimental methods for soil texture classification. By utilizing an easily accessible tool, significant time and cost savings can be expected compared to traditional methods.

• Journal of Broadcast Engineering
• /
• 제18권1호
• /
• pp.21-30
• /
• 2013

It is a challenging problem to most of the image processing when the light source is unknown. The color of the light source must be estimated in order to compensate color changes. To estimate the color of the light source, additional assumption is need, so that we assumed color distribution according to the light source. If the pixels, which do not satisfy the assumption, are used, the estimation fails to provide an accurate result. The most popular color distribution assumption is Grey-World Assumption (GWA); it is the assumption that the color in each scene, the surface reflectance averages to gray or achromatic color over the entire images. In this paper, we analyze the characteristics of the camera response function, and the effect of the Grey-World Assumption on the pixel value and chromaticity, based on the inherent characteristics of the light source. Besides, we propose a novel method that detects important pixels for the color estimation of the light source. In our method, we firstly proposed a method that gives weights to pixels satisfying the assumption. Then, we proposed a pixel detection method, which we modified max-RGB method, to apply on the weighted pixels. Maximum weighted pixels in the column direction and row direction in one channel are detected. The performance of our method is verified through demonstrations in several real scenes. Proposed method better accurately estimate the color of the light than previous methods.

• Journal of the Korea institute for structural maintenance and inspection
• /
• 제23권3호
• /
• pp.83-91
• /
• 2019

The purpose of this study is to assess a fire-damaged concrete structure using a digital camera and image processing software. To simulate it, mortar and paste samples of W/C=0.5(general strength) and 0.3(high strength) were put into an electric furnace and simulated from $100^{\circ}C$ to $1000^{\circ}C$. Here, the paste was processed into a powder to measure CIELAB chromaticity, and the samples were taken with a digital camera. The RGB chromaticity was measured by color intensity analyzer software. As a result, the residual compressive strength of W/C=0.5 and 0.3 was 87.2 % and 86.7 % at the heating temperature of $400^{\circ}C$. However there was a sudden decrease in strength at the temperature above $500^{\circ}C$, while the residual compressive strength of W/C=0.5 and 0.3 was 55.2 % and 51.9 % of residual strength. At the temperature $700^{\circ}C$ or higher, W/C=0.5 and W/C=0.3 show 26.3% and 27.8% of residual strength, so that the durability of the structure could not be secured. The results of $L^*a^*b$ color analysis show that $b^*$ increases rapidly after $700^{\circ}C$. It is analyzed that the intensity of yellow becomes strong after $700^{\circ}C$. Further, the RGB analysis found that the histogram kurtosis and frequency of Red and Green increases after $700^{\circ}C$. It is analyzed that number of Red and Green pixels are increased. Therefore, it is deemed possible to estimate the degree of damage by checking the change in yellow($b^*$ or R+G) when analyzing the chromaticity of the fire-damaged concrete structures.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• 제22권4호
• /
• pp.621-628
• /
• 2021

This paper proposes a method to calibrate a thermal camera with three different perspectives. In particular, the intrinsic parameters of the camera and re-projection errors were provided to quantify the accuracy of the calibration result. Three lenses of the camera capture the same image, but they are not overlapped, and the image resolution is worse than the one captured by the RGB camera. In computer vision, camera calibration is one of the most important and fundamental tasks to calculate the distance between camera (s) and a target object or the three-dimensional (3D) coordinates of a point in a 3D object. Once calibration is complete, the intrinsic and the extrinsic parameters of the camera(s) are provided. The intrinsic parameters are composed of the focal length, skewness factor, and principal points, and the extrinsic parameters are composed of the relative rotation and translation of the camera(s). This study estimated the intrinsic parameters of thermal cameras that have three lenses of different perspectives. In particular, image enhancement based on a deep learning algorithm was carried out to improve the quality of the calibration results. Experimental results are provided to substantiate the proposed method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• 제8권5호
• /
• pp.1080-1088
• /
• 2004

In this paper, the interesting moving objects tracking algorithm using color information on Multi-Video camera against variance of intensity, shape and background is proposed. Moving objects are detected by using difference image method and integral projection method to background image and objects image only with hue area, after converting RGB color coordination of image which is input from multi-video camera into HSI color coordination. Hue information of the detected moving area are segmented to 24 levels from $0^{\circ}$ to $360^{\circ}$. It is used to the feature parameter of the moving objects that are three segmented hue levels with the highest distribution and difference among three segmented hue levels. To examine propriety of the proposed method, human images with variance of intensity and shape and human images with variance of intensity, shape and background are targeted for moving objects. As surveillance results of the interesting human, hue distribution level variation of the detected interesting human at each camera is under 2 level, and it is confirmed that the interesting human is tracked and surveilled by using feature parameters at cameras, automatically.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• 제47권6호
• /
• pp.127-136
• /
• 2010

In multi-view video, illumination disharmony between neighboring views can occur on account of different location of each camera and imperfect camera calibration, and so on. Such discrepancy can be the cause of the performance decrease of multi-view video coding by mismatch of inter-view prediction which refer to the pictures obtained from the neighboring views at the same time. In this paper, we propose an efficient histogram-based prefiltering algorithm to compensate mismatches between the luminance and chrominance components in multi-view video for improving its coding efficiency. To compensate illumination variation efficiently, all camera frames of a multi-view sequence are adjusted to a predefined reference through the histogram matching. A Cosited filter that is used for chroma subsampling in many video encoding schemes is applied to each color component prior to histogram matching to improve its performance. The histogram matching is carried out in the RGB color space after color space converting from YCbCr color space. The effective color conversion skill that has respect to direction of edge and range of pixel value in an image is employed in the process. Experimental results show that the compression ratio for the proposed algorithm is improved comparing with other methods.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• 제36권6호
• /
• pp.589-599
• /
• 2018

The land cover map has been produced by using satellite and aerial images. However, these two images have the limitations in spatial resolution, and it is difficult to acquire images of a area at desired time because of the influence of clouds. In addition, it is costly and time-consuming that mapping land cover map of a small area used by satellite and aerial images. This study used multispectral camera-based drone to acquire multi-temporal images for orthoimages generation. The efficiency of produced land cover map was evaluated using time series analysis. The results indicated that the proposed method can generated RGB orthoimage and multispectral orthoimage with RMSE (Root Mean Square Error) of ${\pm}10mm$, ${\pm}11mm$, ${\pm}26mm$ and ${\pm}28mm$, ${\pm}27mm$, ${\pm}47mm$ on X, Y, H respectively. The accuracy of the pixel-based and object-based land cover map was analyzed and the results showed that the accuracy and Kappa coefficient of object-based classification were higher than that of pixel-based classification, which were 93.75%, 92.42% on July, 92.50%, 91.20% on October, 92.92%, 91.77% on February, respectively. Moreover, the proposed method can accurately capture the quantitative area change of the object. In summary, the suggest study demonstrated the possibility and efficiency of using multispectral camera-based drone in production of land cover map.