• Journal of Korea Society of Digital Industry and Information Management
• /
• v.19 no.3
• /
• pp.245-251
• /
• 2023

Facial expression recognition plays a significant role in understanding human emotional states. With the advancement of AI and computer vision technologies, extensive research has been conducted in various fields, including improving customer service, medical diagnosis, and assessing learners' understanding in education. In this study, we develop a model that can infer emotions in real-time from a webcam using transfer learning with TensorFlow.js and MobileNet. While existing studies focus on achieving high accuracy using deep learning models, these models often require substantial resources due to their complex structure and computational demands. Consequently, there is a growing interest in developing lightweight deep learning models and transfer learning methods for restricted environments such as web browsers and edge devices. By employing MobileNet as the base model and performing transfer learning, our study develops a deep learning transfer model utilizing JavaScript-based TensorFlow.js, which can predict emotions in real-time using facial input from a webcam. This transfer model provides a foundation for implementing facial expression recognition in resource-constrained environments such as web and mobile applications, enabling its application in various industries.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 1993.10a
• /
• pp.1230-1242
• /
• 1993

In a case of mushroom (Lentinus Edodes L.) , visual features are crucial for grading and the quantitative evaluation of the growth state. The extracted quantitative visual features can be used as a performance index for the drying process control or used for the automatic sorting and grading task. First, primary external features of the front and back sides of mushroom were analyzed. And computer vision based algorithm were developed for the extraction and measurement of those features. An automatic thresholding algorithm , which is the combined type of the window extension and maximum depth finding was developed. Freeman's chain coding was modified by gradually expanding the mask size from 3X3 to 9X9 to preserve the boundary connectivity. According to the side of mushroom determined from the automatic recognition algorithm size thickness, overall shape, and skin texture such as pattern, color (lightness) ,membrane state, and crack were quantified and measured. A portion of t e stalk was also identified and automatically removed , while reconstructing a new boundary using the Overhauser curve formulation . Algorithms applied and developed were coded using MS_C language Ver, 6.0, PC VISION Plus library functions, and VGA graphic function as a menu driven way.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2000.11b
• /
• pp.300-306
• /
• 2000

This paper introduces a new concept of automation for bio-production with tele-operative system. The proposed system showed practical and feasible way of automation for the volatile bio-production process. Based on the proposition, recognition of the job environment with object identification was performed using computer vision system. A man-machine interactive hybrid decision-making, which utilized a concept of tele-operation was proposed to overcome limitations of the capability of computer in image processing and feature extraction from the complex environment image. Identifying watermelons from the outdoor scene of the cultivation field was selected to realize the proposed concept. Identifying watermelon from the camera image of the outdoor cultivation field is very difficult because of the ambiguity among stems, leaves, shades, and especially fruits covered partly by leaves or stems. The analog signal of the outdoor image was captured and transmitted wireless to the host computer by R.F module. The localized window was formed from the outdoor image by pointing to the touch screen. And then a sequence of algorithms to identify the location and size of the watermelon was performed with the local window image. The effect of the light reflectance of fruits, stems, ground, and leaves were also investigated.

• 제어로봇시스템학회:학술대회논문집
• /
• 1998.10a
• /
• pp.95-99
• /
• 1998

We introduce an autonomous flying system using a model-helicopter. A feature of the helicopter is that autonomous flight is realized on the low-cost compact model-helicopter. Our helicopter system is divided into two parts. One is on the helicopter, and the other is on the land. The helicopter is loaded with a vision sensor and an electronic compass including a tilt sensor. The control system on the land monitors the helicopter movement and controls. We firstly introduce the configuration of our helicopter system with a vision sensor and an electronic compass. To determine the 3-D position and posture of helicopter, a technique of image recognition using a monocular image is described based on the idea of the sensor fusion of vision and electronic compass. Finally, we show an experiment result, which we obtained in the hovering. The result shows the effectiveness of our system in the compact model-helicopter.

• The KIPS Transactions:PartB
• /
• v.16B no.5
• /
• pp.347-358
• /
• 2009

In this thesis, we develop a vision based hand motion recognition system using a camera with two rotational motors. Existing systems were implemented using a range camera or multiple cameras and have a limited working area. In contrast, we use an uncalibrated camera and get more wide working area by pan-tilt motion. Given an image sequence provided by the pan-tilt camera, color and pattern information are integrated into a tracking system in order to find the 2D position and direction of the hand. With these pose information, we estimate 3D motion plane on which the gesture motion trajectory from approximately forms. The 3D trajectory of the moving finger tip is projected into the motion plane, so that the resolving power of the linear gesture patterns is enhanced. We have tested the proposed approach in terms of the accuracy of trace angle and the dimension of the working volume.

• Journal of the Korean Society of Industry Convergence
• /
• v.22 no.1
• /
• pp.39-45
• /
• 2019

We propose an active color model based method for tracking motions of multiple human using a networked multiple-camera system in IoT space as a human-robot coexistent system. An IoT space is a space where many intelligent devices, such as computers and sensors(color CCD cameras for example), are distributed. Human beings can be a part of IoT space as well. One of the main goals of IoT space is to assist humans and to do different services for them. In order to be capable of doing that, IoT space must be able to do different human related tasks. One of them is to identify and track multiple objects seamlessly. In the environment where many camera modules are distributed on network, it is important to identify object in order to track it, because different cameras may be needed as object moves throughout the space and IoT space should determine the appropriate one. This paper describes appearance based unknown object tracking with the distributed vision system in IoT space. First, we discuss how object color information is obtained and how the color appearance based model is constructed from this data. Then, we discuss the global color model based on the local color information. The process of learning within global model and the experimental results are also presented.

• The Journal of Korea Robotics Society
• /
• v.8 no.1
• /
• pp.51-57
• /
• 2013

Recently, the number of jellyfish has been rapidly grown because of the global warming, the increase of marine structures, pollution, and etc. The increased jellyfish is a threat to the marine ecosystem and induces a huge damage to fishery industries, seaside power plants, and beach industries. To overcome this problem, a manual jellyfish dissecting device and pump system for jellyfish removal have been developed by researchers. However, the systems need too many human operators and their benefit to cost is not so good. Thus, in this paper, the design, implementation, and experiments of autonomous jellyfish removal robot system, named JEROS, have been presented. The JEROS consists of an unmanned surface vehicle (USV), a device for jellyfish removal, an electrical control system, an autonomous navigation system, and a vision-based jellyfish detection system. The USV was designed as a twin hull-type ship, and a jellyfish removal device consists of a net for gathering jellyfish and a blades-equipped propeller for dissecting jellyfish. The autonomous navigation system starts by generating an efficient path for jellyfish removal when the location of jellyfish is received from a remote server or recognized by a vision system. The location of JEROS is estimated by IMU (Inertial Measurement Unit) and GPS, and jellyfish is eliminated while tracking the path. The performance of the vision-based jellyfish recognition, navigation, and jellyfish removal was demonstrated through field tests in the Masan and Jindong harbors in the southern coast of Korea.

• Journal of Biosystems Engineering
• /
• v.27 no.2
• /
• pp.151-160
• /
• 2002

Cow image processing technique would be useful not only for recognizing an individual but also for establishing the image database and analyzing the shape of cows. A cow (Holstein) has usually the unique speckle pattern. In this study, the individual recognition of cow was carried out using the speckle pattern and the content-based image retrieval technique. Sixty cow images of 16 heads were captured under outdoor illumination, which were complicated images due to shadow, obstacles and walking posture of cow. Sixteen images were selected as the reference image for each cow and 44 query images were used for evaluating the efficiency of individual recognition by matching to each reference image. Run-lengths and positions of runs across speckle area were calculated from 40 horizontal line profiles for ROI (region of interest) in a cow body image after 3 passes of 5$\times$5 median filtering. A similarity measure for recognizing cow individuals was calculated using Euclidean distance of normalized G-frame histogram (GH). normalized speckle run-length (BRL), normalized x and y positions (BRX, BRY) of speckle runs. This study evaluated the efficiency of individual recognition of cow using Recall(Success rate) and AVRR(Average rank of relevant images). Success rate of individual recognition was 100% when GH, BRL, BRX and BRY were used as image query indices. It was concluded that the histogram as global property and the information of speckle runs as local properties were good image features for individual recognition and the developed system of individual recognition was reliable.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.5
• /
• pp.187-194
• /
• 2021

A person's behavioral recognition is the recognition of what a person does according to joint movements. To this end, we utilize computer vision tasks that are utilized in image processing. Human behavior recognition is a safety accident response service that combines deep learning and CCTV, and can be applied within the safety management site. Existing studies are relatively lacking in behavioral recognition studies through human joint keypoint extraction by utilizing deep learning. There were also problems that were difficult to manage workers continuously and systematically at safety management sites. In this paper, to address these problems, we propose a method to recognize risk behavior using only joint keypoints and joint motion information. AlphaPose, one of the pose estimation methods, was used to extract joint keypoints in the body part. The extracted joint keypoints were sequentially entered into the Long Short-Term Memory (LSTM) model to be learned with continuous data. After checking the behavioral recognition accuracy, it was confirmed that the accuracy of the "Lying Down" behavioral recognition results was high.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.22 no.7
• /
• pp.107-116
• /
• 2014

This paper presents camera-based road surface marking detection methods suited to sensor fusion-based positioning system that consists of low-cost GPS (Global Positioning System), INS (Inertial Navigation System), EDM (Extended Digital Map), and vision system. The proposed vision system consists of two parts: lane marking detection and RSM (Road Surface Marking) detection. The lane marking detection provides ROIs (Region of Interest) that are highly likely to contain RSM. The RSM detection generates candidates in the regions and classifies their types. The proposed system focuses on detecting RSM without false detections and performing real time operation. In order to ensure real time operation, the gating varies for lane marking detection and changes detection methods according to the FSM (Finite State Machine) about the driving situation. Also, a single template matching is used to extract features for both lane marking detection and RSM detection, and it is efficiently implemented by horizontal integral image. Further, multiple step verification is performed to minimize false detections.