• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권10호
• /
• 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.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권4호
• /
• pp.1927-1943
• /
• 2016

Jitterbug is a passive network covert timing channel supplying reliable stealthy transmission. It is also the basic manner of some improved covert timing channels designed for higher undetectability. The existing entropy-based detection scheme based on training sample binning may suffer from model mismatching, which results in detection performance deterioration. In this paper, a new detection method based on the feature of Jitterbug covert channel traffic is proposed. A fixed binning strategy without training samples is used to obtain bins distribution feature. Coefficient of variation (CV) is calculated for several sets of selected bins and the weighted mean is used to calculate the final CV value to distinguish Jitterbug from normal traffic. Furthermore, the timing window parameter of Jitterbug is estimated based on the detected traffic. Experimental results show that the proposed detection method can achieve high detection performance even with interference of network jitter, and the parameter estimation method can provide accurate values after accumulating plenty of detected samples.

• 반도체디스플레이기술학회지
• /
• 제20권1호
• /
• pp.99-103
• /
• 2021

Warpage defect detecting of scaffold is very important in biosensor production. Because warpaged scaffold cause problem in cell culture. Currently, there is no detection equipment to warpaged scaffold. In this paper, we produced detection model for shape warpage detection using deep learning based CNN. We confirmed the shape of the scaffold that is widely used in cell culture. We produced scaffold specimens, which are widely used in biosensor fabrications. Then, the scaffold specimens were photographed to collect image data necessary for model manufacturing. We produced the detecting model of scaffold warpage defect using Densenet among CNN models. We evaluated the accuracy of the defect detection model with mAP, which evaluates the detection accuracy of deep learning. As a result of model evaluating, it was confirmed that the defect detection accuracy of the scaffold was more than 95%.

• Structural Engineering and Mechanics
• /
• 제80권5호
• /
• pp.563-583
• /
• 2021

Composite materials, composed of multiple constituent materials with dissimilar properties, are actively adopted in a wide range of industrial sectors due to their remarkable strength-to-weight and stiffness-to-weight ratio. Nevertheless, the failure mechanism of composite materials is highly complicated due to their sophisticated microstructure, making it much harder to predict their residual material lives in real life applications. A promising solution for this safety issue is structural damage detection. In the present paper, damage detection of composite material via electrical resistance-based technique and infrared thermography is reviewed. The operating principles of the two damage detection methodologies are introduced, and some research advances of each techniques are covered. The advancement of IR thermography-based non-destructive technique (NDT) including optical thermography, laser thermography and eddy current thermography will be reported, as well as the electrical impedance tomography (EIT) which is a technology increasingly drawing attentions in the field of electrical resistance-based damage detection. A brief comparison of the two methodologies based on each of their strengths and limitations is carried out, and a recent research update regarding the coupling of the two techniques for improved damage detection in composite materials will be discussed.

• Nuclear Engineering and Technology
• /
• 제56권7호
• /
• pp.2799-2805
• /
• 2024

Energy-resolved neutron imaging is an effective way to investigate the internal structure and residual stress of materials. Different sample sizes have varying requirements for the detector's imaging field of view (FOV) and spatial resolution. Therefore, a dual-mode energy-resolved neutron imaging detector was developed, which mainly consisted of a neutron scintillator screen, a mirror, imaging lenses, and a time-stamping optical fast camera. This detector could operate in a large FOV mode or a high spatial resolution mode. To evaluate the performance of the detector, the neutron wavelength spectra and the multiple spatial resolution tests were conducted at CSNS. The results demonstrated that the detector accurately measured the neutron wavelength spectra selected by a bandwidth chopper. The best spatial resolution was about 20 ㎛ in high spatial resolution mode after event reconstruction, and a FOV of 45.0 mm × 45.0 mm was obtained in large FOV mode. The feasibility was validated to change the spatial resolution and FOV by replacing the scintillator screen and adjusting the lens magnification.

• 반도체디스플레이기술학회지
• /
• 제17권4호
• /
• pp.105-108
• /
• 2018

Face Detection is an essential first step of the face recognition, and this is significant effects on face feature extraction and the effects of face recognition. Face detection has extensive research value and significance. In this paper, we present and analysis the principle, merits and demerits of the classic AdaBoost face detection and ASM algorithm based on point distribution model, which ASM solves the problems of face detection based on AdaBoost. First, the implemented scheme uses AdaBoost algorithm to detect original face from input images or video stream. Then, it uses ASM algorithm converges, which fit face region detected by AdaBoost to detect faces more accurately. Finally, it cuts out the specified size of the facial region on the basis of the positioning coordinates of eyes. The experimental result shows that the method can detect face rapidly and precisely, with a strong robustness.

• 한국군사과학기술학회지
• /
• 제24권4호
• /
• pp.392-400
• /
• 2021

Unlike optical equipment, SAR(Synthetic Aperture Radar) has the advantage of obtaining images in all weather, and object detection in SAR images is an important issue. Generally, deep learning-based object detection was mainly performed in real-valued network using only amplitude of SAR image. Since the SAR image is complex data consist of amplitude and phase data, a complex-valued network is required. In this paper, a complex-valued ResNet network is proposed. SAR image object detection was performed by combining the ROI transformer detector specialized for aerial image detection and the proposed complex-valued ResNet. It was confirmed that higher accuracy was obtained in complex-valued network than in existing real-valued network.

• 한국화재소방학회:학술대회논문집
• /
• 한국화재소방학회 1997년도 International Symposium on Fire Science and Technology
• /
• pp.573-581
• /
• 1997

Much activity is occurring throughout the world with respect to the implementation of new concepts and technology for fire detection and fire suppression applications. Obvious advances include tangible detection and suppression improvements, and also new methods and approaches such as performance based codes and standards. Examples of tangible advances include addressable detection systems, wireless detection technology, halon alternatives, water mist systems, advanced sprinkler technology, and so on. Examples of new approaches and methods include a revitalized focus on disaster planning and the need for a total fire protection plan. The concept of performance based codes and standards for the design and installation fire detection and suppression systems will be explored in detail.

• 반도체디스플레이기술학회지
• /
• 제23권1호
• /
• pp.19-24
• /
• 2024

It is not an exaggeration to say that a single fire causes a lot of damage, so fires are one of the disaster situations that must be alerted as soon as possible. Various technologies have been utilized so far because preventing and detecting fires can never be completely accomplished with individual human efforts. Recently, deep learning technology has been developed, and fire detection systems using object detection neural networks are being actively studied. In this paper, we propose a new fire detection system that improves the previously studied fire detection system. We train the YOLOv8 model using refined datasets through improved labeling methods, derive results, and demonstrate the superiority of the proposed system by comparing it with the results of previous studies.

• ETRI Journal
• /
• 제37권2호
• /
• pp.251-261
• /
• 2015

We present a novel method for real-time automatic license plate detection in high-resolution videos. Although there have been extensive studies of license plate detection since the 1970s, the suggested approaches resulting from such studies have difficulties in processing high-resolution imagery in real-time. Herein, we propose a novel cascade structure, the fastest classifier available, by rejecting false positives most efficiently. Furthermore, we train the classifier using the core patterns of various types of license plates, improving both the computation load and the accuracy of license plate detection. To show its superiority, our approach is compared with other state-of-the-art approaches. In addition, we collected 20,000 images including license plates from real traffic scenes for comprehensive experiments. The results show that our proposed approach significantly reduces the computational load in comparison to the other state-of-the-art approaches, with comparable performance accuracy.