• International journal of advanced smart convergence
• /
• v.9 no.3
• /
• pp.232-238
• /
• 2020

In this paper, we study on aerial objects detection and position estimation algorithm for the safety of UAV that flight in BVLOS. We use the vision sensor and LiDAR to detect objects. We use YOLOv2 architecture based on CNN to detect objects on a 2D image. Additionally we use a clustering method to detect objects on point cloud data acquired from LiDAR. When a single sensor used, detection rate can be degraded in a specific situation depending on the characteristics of sensor. If the result of the detection algorithm using a single sensor is absent or false, we need to complement the detection accuracy. In order to complement the accuracy of detection algorithm based on a single sensor, we use the Kalman filter. And we fused the results of a single sensor to improve detection accuracy. We estimate the 3D position of the object using the pixel position of the object and distance measured to LiDAR. We verified the performance of proposed fusion algorithm by performing the simulation using the Gazebo simulator.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.8
• /
• pp.3946-3965
• /
• 2018

Network anomaly detection in Software Defined Networking, especially the detection of DDoS attack, has been given great attention in recent years. It is convenient to build the Traffic Matrix from a global view in SDN. However, the monitoring and management of high-volume feature-rich traffic in large networks brings significant challenges. In this paper, we propose a moving window Principal Components Analysis based anomaly detection and mitigation approach to map data onto a low-dimensional subspace and keep monitoring the network state in real-time. Once the anomaly is detected, the controller will install the defense flow table rules onto the corresponding data plane switches to mitigate the attack. Furthermore, we evaluate our approach with experiments. The Receiver Operating Characteristic curves show that our approach performs well in both detection probability and false alarm probability compared with the entropy-based approach. In addition, the mitigation effect is impressive that our approach can prevent most of the attacking traffic. At last, we evaluate the overhead of the system, including the detection delay and utilization of CPU, which is not excessive. Our anomaly detection approach is lightweight and effective.

• Journal of the Semiconductor & Display Technology
• /
• v.19 no.3
• /
• pp.93-99
• /
• 2020

A majority of car crash is affected by careless driving that causes extensive economic and social costs, as well as injuries and fatalities. Thus, the research of precise crash detection systems is very significant issues in automotive safety. A lot of crash detection algorithms have been developed, but the coverage of these algorithms has been limited to few scenarios. Road scenes and situations need to be considered in order to expand the scope of a collision detection system to include a variety of collision modes. The proposed algorithm effectively handles the x, y, and z axes of the sensor, while considering time and suggests a method suitable for various real worlds. To reduce nuisance and false crash detection events, the algorithm discriminated between driving mode and parking mode. The performance of the suggested algorithm was evaluated under various scenarios, and it successfully discriminated between driving and parking modes, and it adjusted crash detection events depending on the real scenario. The proposed algorithm is expected to efficiently manage the space and lifespan of the storage device by allowing the vehicle's black box system to store only necessary crash event's videos.

• ETRI Journal
• /
• v.33 no.4
• /
• pp.600-610
• /
• 2011

This paper presents a novel approach to face detection by localizing faces as the goal-specific saliencies in a scene, using the framework of selective visual attention of a human with a particular goal in mind. The proposed approach aims at achieving human-like robustness as well as efficiency in face detection under large scene variations. The key is to establish how the specific knowledge relevant to the goal interacts with the bottom-up process of external visual stimuli for saliency detection. We propose a direct incorporation of the goal-related knowledge into the specification and/or modification of the internal process of a general bottom-up saliency detection framework. More specifically, prior knowledge of the human face, such as its size, skin color, and shape, is directly set to the window size and color signature for computing the center of difference, as well as to modify the importance weight, as a means of transforming into a goal-specific saliency detection. The experimental evaluation shows that the proposed method reaches a detection rate of 93.4% with a false positive rate of 7.1%, indicating the robustness against a wide variation of scale and rotation.

• Journal of IKEEE
• /
• v.20 no.1
• /
• pp.82-88
• /
• 2016

The detection rate of Progressive Probability Hough Transform(PPHT) is decreased when a lot of noise components exist due to an unclear or complex original image although it is quite a good algorithm that detects line segments accurately. In order to solve the problem, we propose an improved line detecting algorithm which is robust to noise components and recovers slightly damaged edges. The proposed algorithm is based on PPHT and traces a line segments by pixel and checks of it is straight. It increases the detection rate by reducing the effect of noise components and by recovering edge patterns within a limited pixel size. The proposed algorithm is applied to a lane detection method and the false positive detection rate is decreased by 30% and the line detection rate is increased by 15%.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.7
• /
• pp.1117-1122
• /
• 2017

Pedestrian detection is a very important and valuable part of artificial intelligence and computer vision. It can be used in various areas for example automatic drive, video analysis and others. Many works have been done for the pedestrian detection. The accuracy of pedestrian detection on multiple pedestrian image has reached high level. It is not easily get more progress now. This paper proposes a new structure based on the idea of HOG and convolutional filters to do the pedestrian detection in single pedestrian image. It can be a method to increase the accuracy depend on the high accuracy in single pedestrian detection. In this paper, we use Multiblock HOG and magnitude of the pixel as the feature and use convolutional filter to do the to extract the feature. And then use NEWFM to be the classifier for training and testing. We use single pedestrian image of the INRIA data set as the data set. The result shows that the Convolutional Multiblock HOG we proposed get better performance which is 0.015 miss rate at 10-4 false positive than the other detection methods for example HOGLBP which is 0.03 miss rate and ChnFtrs which is 0.075 miss rate.

• Journal of Korea Multimedia Society
• /
• v.19 no.8
• /
• pp.1310-1319
• /
• 2016

Face detection is the first step in a wide range of face applications. However, detecting faces in the wild is still a challenging task due to the wide range of variations in pose, scale, and occlusions. Recently, many deep learning methods have been proposed for face detection. However, further improvements are required in the wild. Another important issue to be considered in the face detection is the computational complexity. Current state-of-the-art deep learning methods require a large number of patches to deal with varying scales and the arbitrary image sizes, which result in an increased computational complexity. To reduce the complexity while achieving better detection accuracy, we propose a fully convolutional network-based face detection that can take arbitrarily-sized input and produce feature maps (heat maps) corresponding to the input image size. To deal with the various face scales, a multi-scale network architecture that utilizes the facial components when learning the feature maps is proposed. On top of it, we design multi-task learning technique to improve detection performance. Extensive experiments have been conducted on the FDDB dataset. The experimental results show that the proposed method outperforms state-of-the-art methods with the accuracy of 82.33% at 517 false alarms, while improving computational efficiency significantly.

• Korean Journal of Remote Sensing
• /
• v.24 no.5
• /
• pp.445-452
• /
• 2008

Preliminary results are reported on ship detection using coherence images computed from cross-correlating images of multi-look-processed dual-polarization data (HH and HV) of ENVISAT ASAR. The traditional techniques of ship detection by radars such as CFAR (Constant False Alarm Rate) rely on the amplitude data, and therefore the detection tends to become difficult when the amplitudes of ships images are at similar level as the mean amplitude of surrounding sea clutter. The proposed method utilizes the property that the multi-look images of ships are correlated with each other. Because the inter-look images of sea surface are covered by uncorrelated speckle, cross-correlation of multi-look images yields the different degrees of coherence between the images and water. In this paper, the polarimetric information of ships, land and intertidal zone are first compared based on the cross-correlation between HH and HV images, In the next step, we examine the technique when the dual-polarization data are split into two multi-look images, It was shown that the inter-look cross-correlation method could be applicable in the performance improvement of small ship detection and the land masking, It was also found that a simple combination of coherence images from each co-polarised (HH) inter-look and cross-polarised (HV) inter-look data can provide much higher target-detection possibilities.

• Clinical Endoscopy
• /
• v.57 no.2
• /
• pp.217-225
• /
• 2024

Background/Aims: This study aims to compare polyp detection performance of "Deep-GI," a newly developed artificial intelligence (AI) model, to a previously validated AI model computer-aided polyp detection (CADe) using various false positive (FP) thresholds and determining the best threshold for each model. Methods: Colonoscopy videos were collected prospectively and reviewed by three expert endoscopists (gold standard), trainees, CADe (CAD EYE; Fujifilm Corp.), and Deep-GI. Polyp detection sensitivity (PDS), polyp miss rates (PMR), and false-positive alarm rates (FPR) were compared among the three groups using different FP thresholds for the duration of bounding boxes appearing on the screen. Results: In total, 170 colonoscopy videos were used in this study. Deep-GI showed the highest PDS (99.4% vs. 85.4% vs. 66.7%, p<0.01) and the lowest PMR (0.6% vs. 14.6% vs. 33.3%, p<0.01) when compared to CADe and trainees, respectively. Compared to CADe, Deep-GI demonstrated lower FPR at FP thresholds of ≥0.5 (12.1 vs. 22.4) and ≥1 second (4.4 vs. 6.8) (both p<0.05). However, when the threshold was raised to ≥1.5 seconds, the FPR became comparable (2 vs. 2.4, p=0.3), while the PMR increased from 2% to 10%. Conclusions: Compared to CADe, Deep-GI demonstrated a higher PDS with significantly lower FPR at ≥0.5- and ≥1-second thresholds. At the ≥1.5-second threshold, both systems showed comparable FPR with increased PMR.

• IEIE Transactions on Smart Processing and Computing
• /
• v.2 no.6
• /
• pp.357-366
• /
• 2013

Spectrum sensing, as a fundamental functionality of Cognitive Radio (CR), enables Secondary Users (SUs) to monitor the spectrum and detect spectrum holes that could be used. Recently, the security issues of Cognitive Radio Networks (CRNs) have attracted increasing research attention. As one of the attacks against CRNs, a Primary User Emulation (PUE) attack compromises the spectrum sensing of CR, where an attacker monopolizes the spectrum holes by impersonating the Primary User (PU) to prevent SUs from accessing the idle frequency bands. Energy detection is often used to sense the spectrum in CRNs, but the presence of PUE attack has not been considered. This study examined the effect of PUE attack on the performance of energy detection-based spectrum sensing technique. In the proposed protocol, the stationary helper nodes (HNs) are deployed in multiple stages and distributed over the coverage area of the PUs to deliver spectrum status information to the next stage of HNs and to SUs. On the other hand, the first stage of HNs is also responsible for inferring the existence of the PU based on the energy detection technique. In addition, this system provides the detection threshold under the constraints imposed on the probabilities of a miss detection and false alarm.