• Journal of IKEEE
• /
• v.21 no.4
• /
• pp.388-396
• /
• 2017

Since there are many variables such as various poses, illuminations and occlusions in a face detection problem, a high performance detection system is required. Although CNN is excellent in image classification, CNN operatioin requires high-performance hardware resources. But low cost low power environments are essential for small and mobile systems. So in this paper, the CPU-FPGA integrated system is designed based on 3-stage cascade CNN architecture using small size FPGA. Adaptive Region of Interest (ROI) is applied to reduce the number of CNN operations using face information of the previous frame. We use a Field Programmable Gate Array(FPGA) to accelerate the CNN computations. The accelerator reads multiple featuremap at once on the FPGA and performs a Multiply-Accumulate (MAC) operation in parallel for convolution operation. The system is implemented on Altera Cyclone V FPGA in which ARM Cortex A-9 and on-chip SRAM are embedded. The system runs at 30FPS with HD resolution input images. The CPU-FPGA integrated system showed 8.5 times of the power efficiency compared to systems using CPU only.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.3
• /
• pp.112-118
• /
• 2009

Skin color segmentation techniques have been widely utilized for face/hand detection and tracking in many applications such as a diagnosis system using facial information, human-robot interaction, an image retrieval system. In case of a video image, it is common that the skin color model for a target is updated every frame for the robust target tracking against illumination change. As for a single image, however, most of studies employ a fixed skin color model which may result in low detection rate or high false positive errors. In this paper, we propose a novel method for effective skin color segmentation in a single image, which modifies the conditions for skin color segmentation iteratively by the image feedback of segmented skin color region in a given image.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.3
• /
• pp.81-88
• /
• 2012

In this paper, we propose a novel speech enhancement algorithm for effective noise suppression in various noisy environments. In the proposed method, to result in improved decision performance for speech and noise segments, local speech absence probability (LSAP, local SAP) based on Teager energy of noisy speech is used as the feature parameter for voice activity detection (VAD) in each frequency subband instead of conventional LSAP. In addition, The presented method utilizes global SAP (GSAP) derived in each frame as the weighting parameter for the modification of the adopted TE operator to improve the performance of TE operator. Performances of the proposed algorithm are evaluated by objective test under various environments and better results compared with the conventional methods are obtained.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.1
• /
• pp.70-79
• /
• 2017

This study proposes a method to quickly and accurately acquire distance information on direction signs. The proposed method is composed of the recognition of the sign, pre-processing to facilitate the acquisition of the road sign distance, and the acquisition of the distance data. The road sign recognition uses color detection including gamma correction in order to mitigate various noise issues. In order to facilitate the acquisition of distance data, this study applied tilt correction using linear factors, and resolution correction using Fourier transform. To acquire the distance data, morphological operation was used to highlight the area, along with labeling and template matching. By acquiring the distance information on the direction sign through such a processes, the proposed system can be output the distance remaining to the next junction. As a result, when the proposed method is applied to system it can process the data in real-time using the fast calculation speed, average speed was shown to be 0.46 second per frame, with accuracy of 0.65 in similarity value.

• Journal of KIISE
• /
• v.43 no.12
• /
• pp.1412-1419
• /
• 2016

In cognitive radio sensor networks (CRSNs), secondary users (SUs) can occupy licensed bands opportunistically without causing interferences to primary users (PUs). SUs perform spectrum sensing to detect the presence of PUs. Sensing time is a critical parameter for spectrum sensing that can yield a tradeoff between sensing performance and secondary throughput. In this study, we investigate new approaches for spectrum sensing by exploring the tradeoff from a) spectrum sensing for PU detection (SSPD) and b) spectrum sensing for secondary throughput (SSST). In the proposed scheme, the first sensing result of the current frame determines the dynamic performance of the second spectrum sensing. Energy constraint in CRSNs leads to maximized network energy efficiency via optimization of sensing time. Simulation results show that the proposed scheme of SSPD and SSST improves network performance in terms of energy efficiency and secondary throughput, respectively.

• Journal of KIISE:Software and Applications
• /
• v.34 no.12
• /
• pp.1075-1082
• /
• 2007

In this paper, we propose the method of multiple human action recognition on video clip. For being invariant to the change of speed or size of actions, Spatial-Temporal Pyramid method is applied. Proposed method can minimize the complexity of the procedures owing to select Motion Gradient Histogram (MGH) based on statistical approach for action representation feature. For multiple action detection, Motion Energy Image (MEI) of binary frame difference accumulations is adapted and then we detect each action of which area is represented by MGH. The action MGH should be compared with pre-learning MGH having pyramid method. As a result, recognition can be done by the analyze between action MGH and pre-learning MGH. Ten video clips are used for evaluating the proposed method. We have various experiments such as mono action, multiple action, speed and site scale-changes, comparison with previous method. As a result, we can see that proposed method is simple and efficient to recognize multiple human action with stale variations.

• Journal of KIISE
• /
• v.43 no.11
• /
• pp.1281-1297
• /
• 2016

In recent years, personal videos have seen a tremendous growth due to the substantial increase in the use of smart devices and networking services in which users create and share video content easily without many restrictions. However, taking both into account would significantly improve event detection performance because videos generally have multiple modalities and the frame data in video varies at different time points. This paper proposes an event detection method. In this method, high-level features are first extracted from multiple modalities in the videos, and the features are rearranged according to time sequence. Then the association of the modalities is learned by means of DNN to produce a personal video event detector. In our proposed method, audio and image data are first synchronized and then extracted. Then, the result is input into GoogLeNet as well as Multi-Layer Perceptron (MLP) to extract high-level features. The results are then re-arranged in time sequence, and every video is processed to extract one feature each for training by means of DNN.

• Korean Journal of Veterinary Research
• /
• v.43 no.1
• /
• pp.129-137
• /
• 2003

Porcine reproductive and respiratory syndrome (PRRS) is characterized by reproductive failures in sows and respiratory problems in piglets. The nucleocapsid(N) protein, encoded by the open reading frame 7 (ORF7) gene, is known to be the most abundant and antigenic protein in PRRS virus. Therefore, it was suggested that the N protein could be a suitable candidate for the detection of PRRS virus-specific antibodies and diagnosis of PRRS. In the present study, the ORF7 gene encoding the N protein was cloned and expressed as a fusion protein with the glutathione S-transferase (GST) in Escherichia coli. The resulting GST-N recombinant protein was used as an antigen for an indirect sandwich enzyme-linked immunosorbent assay (i-ELISA). Expressed GST-N recombinant protein was migrated at 41 kDa and reacted with ORF7-specific monoclonal antibody by Western blotting. In order to increase the specificity of the ELISA for the detection of PRRS virus-specific antibodes, an i-ELISA was developed using an anti-GST antibody as a capture antibody. The sensitivity and specificity of developed i-ELISA were 92% and 96%, respectively. Based on these results, it was suggested that the i-ELISA is a simple and rapid test for screening a large number of swine sera for the anti-PRRS virus antibodies.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.5 s.311
• /
• pp.45-51
• /
• 2006

Recently, the IDS(Intrusion Detection System) using a video camera is an important part of the home security systems which start gaining popularity. However, the video intruder detection has not been widely used in the home surveillance systems due to its unreliable performance in the environment with abrupt illumination change. In this paper, we propose an effective moving edge extraction algerian from a sequence image. The proposed algorithm extracts edge segments from current image and eliminates the background edge segments by matching them with reference edge list, which is updated at every frame, to find the moving edge segments. The test results show that it can detect the contour of moving object in the noisy environment with abrupt illumination change.

• KIEAE Journal
• /
• v.7 no.4
• /
• pp.147-152
• /
• 2007

Numerous non-destructive tests(NDT) to assess the safety of real structures have been developed. System identification(SI) techniques using dynamic responses and behaviors of structural systems become an outstanding issue of researchers. However the conventional SI techniques are identified to be non-practical to the complex and tall buildings, due to limitation of the availability of an accurate data that is magnitude or location of external loads. In most SI approaches, the information on input loading and output responses must be known. In many cases, measuring the input information may take most of the resources, and it is very difficult to accurately measure the input information during actual vibrations of practical importance, e.g., earthquakes, winds, micro seismic tremors, and mechanical vibration. However, the desirability and application potential of SI to real structures could be highly improved if an algorithm is available that can estimate structural parameters based on the response data alone without the input information. Thus a technique to estimate structural properties of building without input measurement data and using limited response is essential in structural health monitoring. In this study, shaking table tests on three-story plane frame steel structures were performed. Out-put only model analysis on the measured data was performed, and the dynamic properties were inverse analyzed using least square method in time domain. In results damage detection was performed in each member level, which was performed at story level in conventional SI techniques of frequency domain.