• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.9 no.11
• /
• pp.1133-1144
• /
• 2019

Recently, the amount of image on the Internet has rapidly increased, due to the advancement of smart devices and various approaches to effective image retrieval have been researched under these situation. Existing image retrieval methods simply detect the objects in a image and carry out image retrieval based on the label of each object. Therefore, the semantic gap occurs between the image desired by a user and the image obtained from the retrieval result. To reduce the semantic gap in image retrievals, we connect the module for multiple objects classification based on deep learning with the module for human behavior classification. And we combine the connected modules with a behavior ontology. That is to say, we propose an image retrieval system considering the relationship between objects by using the combination of deep learning and behavior ontology. We analyzed the experiment results using walking and running data to take into account dynamic behaviors in images. The proposed method can be extended to the study of automatic annotation generation of images that can improve the accuracy of image retrieval results.

• Journal of radiological science and technology
• /
• v.42 no.1
• /
• pp.39-46
• /
• 2019

This study conducted a comparative analysis of differences between cartesian trajectory in a linear rectangular coordinate system and MultiVane trajectory in a nonlinear rectangular coordinate system axial T1 and axial T2 images using an American College of Radiology(ACR) phantom. The phantom was placed at the center of the head coil and the top-to-bottom and left-to-right levels were adjusted by using a level. The experiment was performed according to the Phantom Test Guidance provided by the ACR, and sagittal localizer images were obtained. As shown in Figure 2, slices # 1 and # 11 were scanned after placing them at the center of a $45^{\circ}$ wedge shape, and a total of 11 slices were obtained. According to the evaluation results, the image intensity uniformity(IIU) was 93.34% for the cartesian trajectory, and 93.19% for the MultiVane trajectory, both of which fall under the normal range in the axial T1 image. The IIU for the cartesian trajectory was 0.15% higher than that for the MultiVane trajectory. In axial T2, the IIU was 96.44% for the cartesian trajectory, and 95.97% for the MultiVane trajectory, which fall under the normal range. The IIU for the cartesian trajectory was by 0.47% higher than that for the MultiVane trajectory. As a result, the cartesian technique was superior to the MultiVane technique in terms of the high-contrast spatial resolution, image intensity uniformity, and low-contrast object detectability.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.4
• /
• pp.171-178
• /
• 2019

The use of anti-scatter grids in radiographic imaging has the advantage of preventing the image distortion caused by scattered radiation. However, it carries the side effect of leaving artifacts in the X-ray image. In this paper, we propose a grid line suppression technique using discrete cosine transform(DCT). In X-ray images, the grid lines have different characteristics depending on the shape of the object and the area of the image. To solve this problem, we adopt the DCT transform based on a dynamic segmentation, and propose a filter transfer function for each individual segment. An algorithm for detecting the band of grid lines in frequency domain and a band stop filter(BSF) with a filter transfer function of a combination of Kaiser window and Butterworth filter have been proposed. To solve the blocking effects, we present a method to determine the pixel values using multiple structured images. The validity of the proposed theory has been evaluated from the experimental results using 140 X-ray images.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.14 no.5
• /
• pp.969-976
• /
• 2019

A lightweight artificial intelligence hardware has made great strides in many application areas. In general, a lightweight artificial intelligence system consist of lightweight artificial intelligence engine and preprocessor including feature selection, generation, extraction, and normalization. In order to achieve optimal performance in broad range of applications, lightweight artificial intelligence system needs to choose a good preprocessing function and set their respective hyper-parameters. This paper proposes a unified framework for a lightweight artificial intelligence system and utilization method for finding models with optimal performance to use on a given dataset. The proposed unified framework can easily generate a model combined with preprocessing functions and lightweight artificial intelligence engine. In performance evaluation using handwritten image dataset and fall detection dataset measured with inertial sensor, the proposed unified framework showed building optimal artificial intelligence models with over 90% test accuracy.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.14 no.6
• /
• pp.15-29
• /
• 2018

This paper proposes an enhanced object tracking algorithm to compensate the lack of temporal information in existing particle swarm optimization based object trackers using the trajectory of the target object. The proposed scheme also enables the tracking and documentation of the location of an online updated set of distractions. Based on the trajectories information and the distraction set, a rule based approach with adaptive parameters is utilized for occlusion detection and determination of the target position. Compare to existing algorithms, the proposed approach provides more comprehensive use of available information and does not require manual adjustment of threshold values. Moreover, an effective weight adjustment function is proposed to alleviate the diversity loss and pre-mature convergence problem in particle swarm optimization. The proposed weight function ensures particles to search thoroughly in the frame before convergence to an optimum solution. In the existence of multiple objects with similar feature composition, this algorithm is tested to significantly reduce convergence to nearby distractions compared to the other existing swarm intelligence based object trackers.

• Journal of the Institute of Convergence Signal Processing
• /
• v.21 no.2
• /
• pp.73-78
• /
• 2020

Musculoskeletal disease is often caused by sitting down for long period's time or by bad posture habits. In order to prevent musculoskeletal disease in daily life, it is the most important to correct the bad sitting posture to the right one through real-time monitoring. In this study, to detect the sitting information of user's without any constraints, we propose posture measurement system based on multi-channel pressure sensor and CNN model for classifying sitting posture types. The proposed CNN model can analyze 5 types of sitting postures based on sitting posture information. For the performance assessment of posture classification CNN model through field test, the accuracy, recall, precision, and F1 of the classification results were checked with 10 subjects. As the experiment results, 99.84% of accuracy, 99.6% of recall, 99.6% of precision, and 99.6% of F1 were verified.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.13 no.6
• /
• pp.545-555
• /
• 2020

This paper presented efficient transceiver structure using sub-band processing for underwater communication in terms of covertness and performance improvement. In aspect of covertness, encrypted coded-bits are divided into groups, and center frequency and sub band are determined by coded-bits of each group. Therefore, as center frequencies are changed randomly, it maintain the covertness effectively. In aspect of performance improvement, the performance of underwater communication mainly depends on multi-path propagation characteristics, Doppler-spread, and frame synchronization. Accordingly, in order to overcome these effects, non-coherent energy detector and turbo equalization method are employed in receiver side. Furthermore, optimal frame synchronization was proposed. Through the simulation and lake experiment, performance analysis was conducted. Especially in the lake experiment, as a result of applying optimal frame synchronization method to receiver structure, errors are corrected in most frames.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.50 no.8
• /
• pp.541-550
• /
• 2022

An effective real time fault diagnosis approach for UAV engine is drawn from IMM filter and GLRT methods. For this purpose based on the linear diagnosis model derived from engine dynamic performance analysis the Kalman filter for residual estimation and each method are applied to the fault diagosis of the actuator for engine control sensors. From the process of the IMM filter application the effective FDI measure is obtained and the state responses due to actuator fault are estimated. Likewise from the GLRT method the fault magnitudes of actuator and sensors are estimated associated with some FDI functionings. The numerical simulations verify the effectiveness of the IMM filter for FDI and the GLRT in estimating the fault magnitudes of each fault mode.

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.5
• /
• pp.25-36
• /
• 2022

This paper proposes a random sign reversal technique in space frequency block code for single carrier modulation. The traditional space time and frequency block coding technique may be confronted with radio environments openly, severe radio hijacking problems are to be overcome. In order to avoid such an open radio issue, random coded data protection technique for space-time block code was proposed, but this algorithm can change channel combination per an Orthogonal Frequency Division Multiplexing block. This kind of slow switching increases the probability that nearby receivers will detect the transmitted data. This paper proposes a fast switching algorithm per data symbols' basis which is a random sign reversal technique in space frequency block code for Single Carrier Modulation. It is shown in simulation that the proposed one has a superior performance in comparison with the performance of the receiver which do not know the random timing sequence of sign reversal.

• Journal of IKEEE
• /
• v.26 no.4
• /
• pp.714-721
• /
• 2022

A phase-locked loop (PLL)-based frequency synthesizer is proposed for a system on a chip (SoC) using multi-frequency clock signals. The proposed PLL-based frequency synthesizer consists of a charge pump PLL which is implemented by a phase frequency detector (PFD), a charge pump (CP), a loop filter, a voltage controlled oscillator (VCO), and a frequency divider, and an edge combiner. The PLL outputs a 12-phase clock by a VCO using six differential delay cells. The edge combiner synthesizes the frequency of the output clock through edge combining and frequency division of the 12-phase output clock of the PLL. The proposed PLL-based frequency synthesizer is designed using a 55-nm CMOS process with a 1.2-V supply voltage. It outputs three clocks with frequencies of 166 MHz, 83 MHz and 124.5MHz for a reference clock with a frequency of 20.75 MHz.