• Journal of Information Processing Systems
• /
• v.19 no.5
• /
• pp.663-672
• /
• 2023

Most vehicle detection methods have poor vehicle feature extraction performance at night, and their robustness is reduced; hence, this study proposes a night vehicle detection method based on style transfer image enhancement. First, a style transfer model is constructed using cycle generative adversarial networks (cycleGANs). The daytime data in the BDD100K dataset were converted into nighttime data to form a style dataset. The dataset was then divided using its labels. Finally, based on a YOLOv5s network, a nighttime vehicle image is detected for the reliable recognition of vehicle information in a complex environment. The experimental results of the proposed method based on the BDD100K dataset show that the transferred night vehicle images are clear and meet the requirements. The precision, recall, mAP@.5, and mAP@.5:.95 reached 0.696, 0.292, 0.761, and 0.454, respectively.

• Journal of Korea Multimedia Society
• /
• v.22 no.1
• /
• pp.44-54
• /
• 2019

Recently, image-based object detection has made great progress with the introduction of Convolutional Neural Network (CNN). Many trials such as Region-based CNN, Fast R-CNN, and Faster R-CNN, have been proposed for achieving better performance in object detection. YOLO has showed the best performance under consideration of both accuracy and computational complexity. However, these data-driven detection methods including YOLO have the fundamental problem is that they can not guarantee the good performance without a large number of training database. In this paper, we propose a data sampling method using CycleGAN to solve this problem, which can convert styles while retaining the characteristics of a given input image. We will generate the insufficient data samples for training more robust object detection without efforts of collecting more database. We make extensive experimental results using the day-time and night-time road images and we validate the proposed method can improve the object detection accuracy of the night-time without training night-time object databases, because we converts the day-time training images into the synthesized night-time images and we train the detection model with the real day-time images and the synthesized night-time images.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.5
• /
• pp.471-478
• /
• 2011

This UGV (Unmanned Ground Vehicle) is not only widely used in various practical applications but is also currently being researched in many disciplines. In particular, obstacle avoidance is considered one of the most important technologies in the navigation of an unmanned vehicle. In this paper, we introduce a simple algorithm for path planning in order to reach a destination while avoiding a polygonal-shaped static obstacle. To effectively avoid such an obstacle, a path planned near the obstacle is much shorter than a path planned far from the obstacle, on the condition that both paths guarantee that the robot will not collide with the obstacle. So, to generate a path near the obstacle, we have developed an algorithm that combines an edge detection method and a limit-cycle navigation method. The edge detection method, based on Hough Transform and IR sensors, finds an obstacle's edge, and the limit-cycle navigation method generates a path that is smooth enough to reach a detected obstacle's edge. And we proposed novel algorithm to solve local minima using the virtual wall in the local vision. Finally, we verify performances of the proposed algorithm through simulations and experiments.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.149-154
• /
• 2024

This paper proposes an algorithm that remedy Floyd's the tortoise and the hare algorithm (THA) shortcomings which is specialized in singly linked list (SLL), so this algorithm fails to detect the cycle in undirected graph, digraph, and tree with multiple inputs or outputs. The proposed algorithm simply pruning the source and sink with only one edge using cycle detection of single edge node pruning. As a result of the experimental of various list, undirected graph, digraph, and tree, the proposed algorithm can be successively detect the cycle all of them. Thus, the proposed algorithm has the simplest and fastest advantage in the field of cycle detection.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.8 no.6
• /
• pp.1-8
• /
• 2000

A new method of knock detection in SI engines, using a change of ion concentration in the combustion chamber, was developed. In order to measure in-cylinder ionization current, ionization probes were installed at spark plug and cylinder head of production engine. It was found that the electric current generated by ionized gas in core burned gas region of knocking cycle is between 2 and 10 times larger than that of normal cycle, because the burned gas temperature which is the dominant parameter of a change of ion concentration increases. However, a change of ionization current in boundary region of burned gas is relatively weak. Hence a change of ion concentration in core burned gas region can be used for knock detection.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.6
• /
• pp.91-97
• /
• 2022

There is hare and tortoise racing algorithm(HTA) for single-source(SS) singly linked list(SLL) with O(n) time complexity. But the fast method is unknown for general graph with multi-source, multi-destination, and multi-branch(MSMDMB). This paper suggests linear time cycle detection algorithm for given undirected and digraph with MSMDMB. The proposed method reduced the given graph G contained with unnecessary vertices(or nodes) to cycle into reduced graph G' with only necessary vertices(or nodes) to cycle based on the condition of cycle formation. For the reduced graph G', we can be find the cycle set C and cycle length λ using linear search within linear time. As a result of experiment data, the proposed algorithm can be obtained the cycle for whole data.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2004.02a
• /
• pp.82-94
• /
• 2004

This paper addresses solutions to the challenges of carrier phase integer ambiguity resolution and cycle slip detection/identification, for maintaining high accuracy of an integrated GPS/Pseudolite/INS system. Such a hybrid positioning and navigation system is an augmentation of standard GPS/INS systems in localized areas. To achieve the goal of high accuracy, the carrier phase measurements with correctly estimated integer ambiguities must be utilized to update the system integration filter's states. The occurrence of a cycle slip that is undetected is, however, can significantly degrade the filter's performance. This contribution presents an effective approach to increase the reliability and speed of integer ambiguity resolution through using pseudolite and INS measurements, with special emphasis on reducing the ambiguity search space. In addition, an algorithm which can effectively detect and correct the cycle slips is described as well. The algorithm utilizes additional position information provided by the INS, and applies a statistical technique known as the cumulative-sum (CUSUM) test that is very sensitive to abrupt changes of mean values. Results of simulation studies and field tests indicate that the algorithms are performed pretty well, so that the accuracy and performance of the integrated system can be maintained, even if cycle slips exist in the raw GPS measurements.

• Journal of Biomedical Engineering Research
• /
• v.23 no.6
• /
• pp.477-484
• /
• 2002

A pacemaker autosense algorithm with dual thresholds. one for noise or tachyarrhythmia detection (noise threshold, NT) and the other for intrinsic beat detection (sensing threshold. ST), was developed to improve the sensing performance in single pass VDD electrograms. unipolar electrograms, or atrial fibrillation detection. When a deflection in an electrogram exceeds the NT (defined as 50% of 57), the autosense algorithm with dual thresholds checks if the deflection also exceeds the ST. If it does, the autosense algorithm calculates the signal to noise ratio (SNR) of the deflection to the highest deflection detected by NT but lower than ST during the last cardiac cycle. If the SNR 2, the autosense algorithm declares an intrinsic beat detection and calculates the next ST based on the three most recent intrinsic peaks. If the SNR $\geq$2, the autosense algorithm checks the number of deflections detected by NT during the last cardiac cycle in order to determine if it is a noise detection or tachyarrhythmia detection. Usually the autosense algorithm tries to set the 57 at 37.5% of the average of the three intrinsic beats, although it changes the percentage according to event classifications. The autosense algorithm was tested through computer simulation of atrial electrograms from 5 patients obtained during EP study, to simulate a worst sensing situation. The result showed that the ST levels for autosense algorithm tracked the electrogram amplitudes properly, providing more noise immunity whenever necessary. Also, the autosense algorithm with dual thresholds achieved sensing performance as good as the conventional fixed sensitivity method that was optimized retrospectively.

• Journal of Sensor Science and Technology
• /
• v.32 no.2
• /
• pp.67-74
• /
• 2023

A microfluidic paper-based analytical device (µPAD) is proposed for the selective detection of ammonia in water by using the modified Berthelot reagent and a fluidic channel consisting of hollow paper. The modified Berthelot reagents were uniformly dispersed in cyclohexane and then immobilized in a detection zone of the µPAD. The loading position of the reagents and the type of a sample flow channel were optimized to achieve a sensitive ammonia detection within a short analytical time. The NH₃ µPAD exhibits a linear colorimetric response to the concentration of ammonia dissolved in water in the range of 1-100 mg L^-1, and its limit-of-detection is 1.75 mg L^-1. In addition, the colorimetric response was not influenced by the addition of 100 mg L^-1 nitrogen containing compounds (sodium nitrate, sodium nitrite, uric acid, hydroxylamine, butylamine, diethylamine) or inorganic salts (NaCl, Na₂HPO₄), presenting the enough selectivity in the detection of water-dissolved ammonia against possible interferents.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.29 no.6
• /
• pp.1016-1024
• /
• 2000

The study was conducted to develop a rapid method for the detection of Listeria monocytogenes in foods via polymerase chain reaction (PCR) technique using hemolysin gene (hlyA) primers. Specificity and sensitivity of PCR, optimal conditions for PCR and application of hlyA gene primers for the detection of L. monocytogenes from milk and beef were investigeted. Each of the 20 L. monocytogenes strains gave a single 713 bp band, but other Listeria sup. and other bacteria did not show any bands. As few as 1 pg of L. monocytogenes DNA or 2.4$\times$10$^4$L. monocytogenes cells could be detected with hlyA gene primers. PCR product was most improved at 20~30 cycle in terms of removal of tailing and sensitivity. Also, the sensitivity was significantly improved by the further 10~15 cycle after 20 cycle PCR amplication. Milk (10 mL) and beef (10 g) samples were inoculated with L. monocytogenes at the concentrations ranging from 0 to 10$^{7}$ CFU/mL or g to determine the best sensitivity of PCR for the rapid detection of L. monocytogenes. PCR assay could detect 2 cells in milk with repeating PCR amplication and 2.6$\times$10$^2$cells in beef sample after 24 hr enrichment growth at 35$^{\circ}C$ in LEB.