• Smart Media Journal
• /
• v.12 no.9
• /
• pp.159-170
• /
• 2023

It takes a lot of time and manpower to search for the missing. As part of the solution, a missing person search AI system was implemented using a YOLO-based model. In order to train object detection models, the model was learned by collecting recognition images (road fixation) of drone mobile objects from AI-Hub. Additional mountainous terrain datasets were also collected to evaluate performance in training datasets and other environments. In order to optimize the missing person search AI system, performance evaluation based on model size and hyperparameters and additional performance evaluation for concerns about overfitting were conducted. As a result of performance evaluation, it was confirmed that the YOLOv5-L model showed excellent performance, and the performance of the model was further improved by applying data augmentation techniques. Since then, the web service has been applied with the YOLOv5-L model that applies data augmentation techniques to increase the efficiency of searching for missing people.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.1
• /
• pp.15-21
• /
• 2017

In this paper, we present a solution to the problems caused by using only text - based information as an index element when a commercialized missing person indexing system indexes missing persons registered in the database. The existing system could not be used for the missing persons inquiry because it could not formalize the image of the missing person registered together when registering the missing person. To solve these problems, we propose a method to extract the similarity of images by using OpenCV image processing and TensorFlow deep - running image processing, and to process images of missing persons to process them into meaningful information. In order to verify the indexing method used in this paper, we constructed a Web server that operates to provide the information that is most likely to be needed to users first, using the image provided in the non - regular environment of the same subject as the search element.

• Journal of Korea Society of Industrial Information Systems
• /
• v.25 no.3
• /
• pp.53-59
• /
• 2020

This paper proposes color segmentation for detecting missing persons in a smart lighting system using radar and camera sensors. Recently, smart lighting systems built-in radar and cameras have been efficient in saving energy and searching for missing persons, simultaneously. In smart lighting systems, radar detects moving objects and then the lights turn on and camera records. The video recorded is useful to find out missing persons. The color of their clothes worn in missing persons is one of critical hints to look for missing persons. Therefore, color segmentation is an effective means for detecting the color of their clothes. In this paper, during the color segmentation step, the ROI(Region of interest) setting based on the size of an object is applied and the background is reduced. According to experimental results, the color segmentation has good accuracy of more than 97%.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.1
• /
• pp.139-146
• /
• 2022

Mountain accidents such as forest fires and missing people are increasing as hikers increase due to indoor activities restrictions caused by the prolonged COVID-19 incident. If a dangerous situation occurs at outside where rescue workers cannot reach, the search time for person can be reduced using a quadcopter. Considering this, in this paper, Multiwii is used to smoothly hover the quadcopter by setting optimized PID values of the x-axis, y-axis, and z-axis (Yaw) according to the change in the inclination of the gas. In addition, after installing Open CV on Raspberry Pie, the camera uses HSV color space to filter the color such as the description of the person, and uses a thermal imaging camera to receive thermal sensing images in real time in environments where color extraction is difficult. As a result, it was confirmed that hovering was possible at a height of 2 to 8 m, blue extraction was possible at a height of 5 m, and heat detection was possible at a distance of less than 10 cm.

• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.171-187
• /
• 2023

When a person's face is recognized through a recording device such as a low-pixel surveillance camera, it is difficult to capture the face due to low image quality. In situations where it is difficult to recognize a person's face, problems such as not being able to identify a criminal suspect or a missing person may occur. Existing studies on face recognition used refined datasets, so the performance could not be measured in various environments. Therefore, to solve the problem of poor face recognition performance in low-quality images, this paper proposes a method to generate high-quality images by performing image quality improvement on low-quality facial images considering various environments, and then improve the performance of facial feature point detection. To confirm the practical applicability of the proposed architecture, an experiment was conducted by selecting a data set in which people appear relatively small in the entire image. In addition, by choosing a facial image dataset considering the mask-wearing situation, the possibility of expanding to real problems was explored. As a result of measuring the performance of the feature point detection model by improving the image quality of the face image, it was confirmed that the face detection after improvement was enhanced by an average of 3.47 times in the case of images without a mask and 9.92 times in the case of wearing a mask. It was confirmed that the RMSE for facial feature points decreased by an average of 8.49 times when wearing a mask and by an average of 2.02 times when not wearing a mask. Therefore, it was possible to verify the applicability of the proposed method by increasing the recognition rate for facial images captured in low quality through image quality improvement.