• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.4
• /
• pp.349-355
• /
• 2015

Underwater robots generally show better performances for tasks than humans under certain underwater constraints such as. high pressure, limited light, etc. To properly diagnose in an underwater environment using remotely operated underwater vehicles, it is important to keep autonomously its own position and orientation in order to avoid additional control efforts. In this paper, we propose an efficient method to assist in the operation for the various disturbances of a remotely operated vehicle for the diagnosis of underwater structures. The conventional AHRS-based bearing estimation system did not work well due to incorrect measurements caused by the hard-iron effect when the robot is approaching a ferromagnetic structure. To overcome this drawback, we propose a sensor fusion algorithm with the camera and AHRS for estimating the pose of the ROV. However, the image information in the underwater environment is often unreliable and blurred by turbidity or suspended solids. Thus, we suggest an efficient method for fusing the vision sensor and the AHRS with a criterion which is the amount of blur in the image. To evaluate the amount of blur, we adopt two methods: one is the quantification of high frequency components using the power spectrum density analysis of 2D discrete Fourier transformed image, and the other is identifying the blur parameter based on cepstrum analysis. We evaluate the performance of the robustness of the visual odometry and blur estimation methods according to the change of light and distance. We verify that the blur estimation method based on cepstrum analysis shows a better performance through the experiments.

• ETRI Journal
• /
• v.30 no.1
• /
• pp.59-67
• /
• 2008

The necessity for the precise time synchronization of measurement data from multiple sensors is widely recognized in the field of global positioning system/inertial navigation system (GPS/INS) integration. Having precise time synchronization is critical for achieving high data fusion performance. The limitations and advantages of various time synchronization scenarios and existing solutions are investigated in this paper. A criterion for evaluating synchronization accuracy requirements is derived on the basis of a comparison of the Kalman filter innovation series and the platform dynamics. An innovative time synchronization solution using a counter and two latching registers is proposed. The proposed solution has been implemented with off-the-shelf components and tested. The resolution and accuracy analysis shows that the proposed solution can achieve a time synchronization accuracy of 0.1 ms if INS can provide a hard-wired timing signal. A synchronization accuracy of 2 ms was achieved when the test system was used to synchronize a low-grade micro-electromechanical inertial measurement unit (IMU), which has only an RS-232 data output interface.

• Journal of Mechanical Science and Technology
• /
• v.16 no.10
• /
• pp.1253-1263
• /
• 2002

The control system design and modeling of an unmanned vehicle by means of a new concept for better performance through a tole-operation system is suggested by sensor fusion. But, the control of a real vehicle is very difficult, because the system identification of the vehicle is hard to find the unknown factors and the disturbances of the experimental environment. For the longitudinal and lateral controls, the traction system and steering system models are set up and a tuning method to find the gain of the controller by experiments is presented. In this research, mechanical and electronic parts are implemented to operate the unmanned vehicle and data reconstruction method of information about the environment data coming from several sensors is presented by data plot for the vehicle navigation. This paper focuses on the integration of tole-operated unmanned vehicle. This vehicle mainly controlled lateral and longitudinal directions with actuators for controlling vehicle movement and sensors for the closed-loop controlled system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.10
• /
• pp.4080-4097
• /
• 2020

In recent years, convolutional neural network (CNN) has become the primary method for face detection. But its shortcomings are obvious, such as expensive calculation, heavy model, etc. This makes CNN difficult to use on the mobile devices which have limited computing and storage capabilities. Therefore, the design of lightweight CNN for face detection is becoming more and more important with the popularity of smartphones and mobile Internet. Based on the CPU real-time face detector FaceBoxes, we propose a multi-task lightweight face detector, which has low computing cost and higher detection precision. First, to improve the detection capability, the squeeze and excitation modules are used to extract attention between channels. Then, the textual and semantic information are extracted by shallow networks and deep networks respectively to get rich features. Finally, the landmark detection module is used to improve the detection performance for small faces and provide landmark data for face alignment. Experiments on AFW, FDDB, PASCAL, and WIDER FACE datasets show that our algorithm has achieved significant improvement in the mean average precision. Especially, on the WIDER FACE hard validation set, our algorithm outperforms the mean average precision of FaceBoxes by 7.2%. For VGA-resolution images, the running speed of our algorithm can reach 23FPS on a CPU device.

• ETRI Journal
• /
• v.42 no.6
• /
• pp.976-986
• /
• 2020

Internet of Things (IoT) is considered the future network to support wireless communications. To realize an IoT network, sufficient spectrum should be allocated for the rapidly increasing IoT devices. Through cognitive radio, unlicensed IoT devices exploit cooperative spectrum sensing (CSS) to opportunistically access a licensed spectrum without causing harmful interference to licensed primary users (PUs), thereby effectively improving the spectrum utilization. However, an open access cognitive IoT allows abnormal IoT devices to undermine the CSS process. Herein, we first establish a hard-combining attack model according to the malicious behavior of falsifying sensing data. Subsequently, we propose a weighted sequential hypothesis test (WSHT) to increase the PU detection accuracy and decrease the sampling number, which comprises the data transmission status-trust evaluation mechanism, sensing data availability, and sequential hypothesis test. Finally, simulation results show that when various attacks are encountered, the requirements of the WSHT are less than those of the conventional WSHT for a better detection performance.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_1
• /
• pp.1283-1297
• /
• 2023

Mid-wave infrared (MWIR) imagery, due to its ability to capture the temperature of land cover and objects, serves as a crucial data source in various fields including environmental monitoring and defense. The KOMPSAT-3A satellite acquires MWIR imagery with high spatial resolution compared to other satellites. However, the limited spatial resolution of MWIR imagery, in comparison to electro-optical (EO) imagery, constrains the optimal utilization of the KOMPSAT-3A data. This study aims to create a highly visible MWIR fusion image by leveraging the edge information from the KOMPSAT-3A panchromatic (PAN) image. Preprocessing is implemented to mitigate the relative geometric errors between the PAN and MWIR images. Subsequently, we employ a pre-trained pixel difference network (PiDiNet), a deep learning-based edge information extraction technique, to extract the boundaries of objects from the preprocessed PAN images. The MWIR fusion imagery is then generated by emphasizing the brightness value corresponding to the edge information of the PAN image. To evaluate the proposed method, the MWIR fusion images were generated in three different sites. As a result, the boundaries of terrain and objects in the MWIR fusion images were emphasized to provide detailed thermal information of the interest area. Especially, the MWIR fusion image provided the thermal information of objects such as airplanes and ships which are hard to detect in the original MWIR images. This study demonstrated that the proposed method could generate a single image that combines visible details from an EO image and thermal information from an MWIR image, which contributes to increasing the usage of MWIR imagery.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.9
• /
• pp.4665-4683
• /
• 2019

Previous methods build image annotation model by leveraging three basic dependencies: relations between image and label (image/label), between images (image/image) and between labels (label/label). Even though plenty of researches show that multiple dependencies can work jointly to improve annotation performance, different dependencies actually do not "work jointly" in their diagram, whose performance is largely depending on the result predicted by image/label section. To address this problem, we propose the adaptive attention annotation model (AAAM) to associate these dependencies with the prediction path, which is composed of a series of labels (tags) in the order they are detected. In particular, we optimize the prediction path by detecting the relevant labels from the easy-to-detect to the hard-to-detect, which are found using Binary Cross-Entropy (BCE) and Triplet Margin (TM) losses, respectively. Besides, in order to capture the inforamtion of each label, instead of explicitly extracting regional featutres, we propose the self-attention machanism to implicitly enhance the relevant region and restrain those irrelevant. To validate the effective of the model, we conduct experiments on three well-known public datasets, COCO 2014, IAPR TC-12 and NUSWIDE, and achieve better performance than the state-of-the-art methods.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.9
• /
• pp.49-57
• /
• 1999

The fusion of fuzzy set theory and neural networks technologies have concentrated on applying neural networks to obtain the optimal rule bases of fuzzy logic system. Unfortunately, this is very hard to achieve due to limited learning capabilities of neural networks. To overcome this difficulty, we propose a new approach in which rough set theory and neuro-fuzzy fusion are combined to obtain the optimal rule base from input/output data. Compared with conventional FNN, the proposed algorithm is considerably more realistic because it reduces overlapped data when construction a rule base. This results are applied to the construction of inference rules for controlling the temperature at specified points in a refrigerator.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.1 no.1
• /
• pp.22-32
• /
• 2002

This paper describes about non-burial AVI (Automatic Vehicle Identification) system using general vehicle as probe car for obtaining more accurate traffic information while conserving road pavement surface. Existing spot traffic detectors have their own limits of not obtaining right information owing to its mathematical method. Burial AVI systems have some defects, causing traffic jam, needing much maintenance cost because of frequent cutting of loop and piezo-electric sensors. Especially, they have hard time to make right detection, when it comes to jamming time. Therefore, in this paper, we propose non-burial AVI system with laser trigger unit. Proposed non-burial AVI system is developed to obtain regional traffic information from normal Passing vehicle by automatic license number recognition technology. We have adapted it to national highway section between Suwon city and Pyong$\~$Taek city(9.5km) and get affirmative results. Vehicle detection rate of laser trigger unit is more than 95$\%$, vehicle recognition rate is 87.8$\%$ and vehicle matching rate is about 14.3$\%$. So we regard these as satisfying results to use the system for traffic information service. We evaluate proposed AVI system by regulation of some institutions which are using similar AVI system and the proposed system satisfies all conditions. For future study, we have plan of detailed research about proper lane number from all of the target lanes, optimal section length, information service period, and data fusion method for existing spot detector.

• Journal of Internet Computing and Services
• /
• v.12 no.2
• /
• pp.1-8
• /
• 2011

In this paper, we propose an effective method for cooperative spectrum sensing in cognitive radios where cognitive user(CR) with the highest reliability sensing data is only selected and allowed to report its local decision to FC as only decision making node. The proposed scheme enables CR users to implicitly compare their sensing data reliabilities based on their likelihood ratio, without any collaboration among cognitive radio users. Due to the mechanism, the proposed cooperative scheme can achieves a high spectrum sensing performance while only requiring extremely low cooperation resources such as signaling overhead and cooperative time in comparison with other existing methods such as maximum ratio combination (MRC) based, equal gain combination (EGC) based and conventional hard combination based cooperative sensing methods.