|
A study on calibration of a speedometer using GNSS Simulator |
|
Wangheon Lee(Hansei University, Korea), Kwang-Suk Oh(Korea Research Center for Measuring Instruments, Korea) |
| In this study, we conducted the speed correction of car using satellite coordinates as well as speed moving Earth's orbit 24 hours a day. There is defects that the same three-dimensional spatial information cannot be obtained at every orbiting time because the orbit of the satellite is not constant for each rotation. The spatial coordinates circulating around the Earth's orbit are measured in real time and stored in a GNSS simulator. Using this information without driving the car, the vehicle speed can be corrected moving GNSS simulator. After setting Cheorwon Area as the driving area for back box calibration, we confirmed the usefulness of proposed GNSS simulator based black box calibration |
|
|
Comparative Analysis of IR-IR Image Matching Applying the Deep Learning-based Template Matching Techniques |
|
Seungeon Lee, Sungho Kim(Yeungnam University, Korea) |
| Template matching technology is a technique that can be applied to various computer vision technologies such as object detection, object recognition, and object tracking. A powerful template matching technique can improve the accuracy and effectiveness of computer vision techniques. In this paper, the template matching techniques are compared with Infrared Radiation images. The techniques consist of QATM (Quality Aware Template Matching), Deep-DIM (Divisive Input Modulation) and SiamFC. The input IR image pairs consist of a search image and a template image, and the template image is an image that crop out of the search image. Each technique is compared and analyzed for runtime and accuracy. |
|
|
Technology Trend for Compression of Deep Learning Model Algorithm |
|
Saebyeol Do, Kim sungho(Yeungnam University, Korea) |
|
|
One-stage Infrared Ships Detection with Attention Mechanism |
|
Indah Monisa Firdiantika, Sungho Kim(Yeungnam University, Korea) |
| Infrared marine target detection is a key technology in marine monitoring. However, Infrared ship detection still encounters complicated challenges due to the insufficient spatial resolution and low signal-to-noise ratio resulting in an extreme lack of texture details for small infrared objects. At the same time, the ocean environment is changeable, unclear, and complex due to the influence of many disturbances. To alleviate these challenges, a one-stage object detection based on You Only Look Once (YOLOv7) model with attention mechanism is proposed for infrared ship detection. Our approach incorporated the Coordinate Attention for Efficient Mobile Network Design (CA) with YOLOv7 to reduce d |
|
|
A study on development of multicopter with double propeller |
|
JaeJin Jung(Hansei university, Korea) |
| Multicopters are used in various fields such as hobbies, surveillance and reconnaissance, filming, lifesaving, and agricultural control because they have the advantage of relatively simple operation principle, easy control, and low cost such as price and maintenance. Based on these market trends, this paper conducted a study on multicopters equipped with dual propellers as a way to respond to the demand for miniaturization of aircraft. In particular, the propeller was designed in an overlapping structure to make it smaller than the existing multicopters, and as a result of analyzing the efficiency, vibration, and flight stability of the multicopter developed at the same time, it was confirme |
|
|
Development of an R/S/T Terminal Recognition System for Electrical Distribution Panel Operating Robots Using Depth Camera |
|
Choonggun Kim(Sogang University, Korea), Hanwoo Lee, Myoungchul Lee, Sooan Choi(ES-Robot Co., Ltd., Korea), Doyoung Jeon(Sogang University, Korea) |
| In high-voltage distribution panels, electricians follow safety protocols when performing tasks such as inspection, grounding, and mega-testing to prevent hazards. However, accidents occur several times a year, leading to a demand for automating these tasks using robots. A critical step towards this automation is recognizing and locating the R/S/T terminals of the distribution panel. In this study, using a single depth camera and the MASK r-cnn algorithm, we identified the relative x, y, z positions of the terminals with an RMS error of 1.76 mm, and the automation system utilizes this information to perform the required tasks. |
|