Search

Millimeter-wave radar has proven to have a good range estimation accuracy and is less influenced by weather conditions. However, it is difficult for radar to recognize objects, and it is prone to cause a false alarm. In this paper, we present an object detection and classification by jointly using a radar and camera sensors for traffic surveillance applications. The proposed method fuses the Regions of Interest (ROIs) generated on each of the detection results obtained independently from radar and camera sensors. Reducing the high false alarm of a radar sensor is the main aim of the fusion method. Then, a Convolutional Neural Network (CNN) is used to classify the final fused detected objects into one of the six-vehicle categories; Sedan, Truck, Minivan, Bus, Microbus, and SUV. The proposed method was verified using real data. Results obtained demonstrate the good performance of the proposed fusion approach in traffic surveillance context.

In this paper, we briefly introduce the concept of 3D frequency selective rasorber (FSR) as well as its generalized equivalent circuit model. Based on this understanding, a new kind of 3D FSR based on stacked planar slotline structure is presented. As an example, a prototype 3D FSR is demonstrated to exhibit bandpass filtering response and good absorption performance at both sides of the passband, and it is then fabricated and measured. The measured results show that the minimum insertion loss in the transmission window is about 0.3 dB, and the absorption bandwidths are ranged from 1.6 to 2.3 GHz and from 3.6 to 4.6 GHz, respectively. To further verify the design concept, another design example with wider absorption bands is developed by introducing more lossy resonators.

This paper presents an absorptive frequency selective reflector (AFSR), which exhibits a flat reflection characteristic with absorption at the both sides of the reflection band. This structure formed in planar slotline structures is mainly inclusive of one lossless resonator and two lossy resonators, which are responsible for one reflection band and two absorption bands respectively. In order to minimize the coupling of these three resonators, metallic plates are utilized to make these three resonators separated. In this paper, equivalent circuit model (ECM) based on transmission line theory is introduced to describe the design concept. Then, the realization of specified structure is provided, where a good agreement of reflection coefficient results between EM simulation and ECM is achieved. The proposed AFSR features stable performance under oblique incident angles up to 40°.

We have published a ground-coupled GPR antenna array, which consists of one transmitter and five receivers, for in situ estimation of the thickness and dielectric permittivity of an asphalt pavement layer [1]. In this paper, we optimize the design of the antipodal Vivaldi antenna to pursue a higher resolution through reducing the late-time ringing. Furthermore, an off-ground antenna array is assembled, and can effectively accelerate the acquisition of common-source GPR datasets. The antenna phase center was calibrated to estimate the velocity and thickness of both the air gap and asphalt pavement layer, through velocity spectrum analysis. Numerical experimental results show that the off-ground antenna array is one ideal method for inspecting asphalt pavements.

Fast implementation of parameter estimation for polynomial phase signal (PPS) is considered in this paper. A method which combines the least squares unwrapping (LSU) estimator and the extended Kalman filter (EKF) is proposed. A small number of initial samples are used to estimate the PPS's parameters and then these coarse estimates are used to initial the EKF. The proposed LSU-EKF estimator greatly reduces the computation complexity of the LSU estimator and can work in entire identifiable region which inherits from the LSU estimator. Meanwhile, in the EKF stage its output is in point-by-point wise which is useful in real applications.

Depth estimation is a crucial step for 2D/3D conversion from monoscopic video. In this paper, a novel method for depth estimation from motion with camera motion is proposed. In the proposed method, image matching using critical point filters is applied to extract the pixel-level motion field for each frame. As camera motion can bring pseudo motion vectors by image matching, and thus leading to depth ambiguity. To solve this problem, we propose to estimate the camera moving model using robust RANSAC algorithm. Then, the initial depth map is estimated by using the motion vectors without camera motion. Finally, the depth values of the pixels at the edges of moving objects are refined using a post filter based on homogeneous points. Experimental results show that the proposed method achieves considerable performances on depth map in presence of camera motion.

As the innovative products of Office, Office2007 and Office2010 use Open XML language to describe and ZIP to pack, and choose the SHA1-AES encryption mechanism. The paper started from the document structure, extracted the encryption information flow, and then cracked respectively the password on CPU computing computer and GPU computing platform by password brute-force method, analyzed the security of Office at last.

Multiresolutional critical point filters (CPF), a set of nonlinear filters, has been shown to be effective in image matching interpolation. In CPF matching, the elastic model defined by the bijectivity conditions can express deformable motions accurately. But the surjection condition constrains that the pixel on the boundary of the source image should be mapped to the pixel with the same location at the destination image. This does not hold in many cases, and will lead to mapping convergence at the image boundary, and thus degrading the quality of interpolated images. To solve this problem, this paper proposes an unconstrained boundary extension (UBE) approach. By UBE, mapping convergence at the image boundary can be released. Experimental results show that the proposed approach can improve the visual quality of the interpolated images, especially at the boundary regions.

Deblocking filter of H.264 is identified as an important technology for improving the subjective quality of video. However, it is difficult to apply to low-cost embedded terminals due to its high computational complexity. By analyzing the statistical correlations of boundary strength and true-edge detection checks among adjacent blocks and pixels, this paper proposes an enhanced method of deblocking filter that efficiently reduces the deblocking complexity. In the ‘statistics-based deblocking filter (SBDF)’ proposed in this paper, only one boundary strength (BS) derivation is required for each macroblock edge instead of original four BS derivations, and four true-edge detections for each macroblock edge instead of original 16 true-edge detections. Experimental results show that the proposed SBDF significantly reduce the computational complexity with a slight loss in video quality.

The tree structure based block partition in H.264 has been shown to be more efficient than simple uniform block partition. However, tree based partition still does not code efficiently enough with video information, as it is unable to capture the geometric structure of 2D data. Concerned with this problem, this paper proposes a new motion compensation technique using polyline based block partition. A polyline, composed of two line segments, is used to approximate the curve contour of moving objects. By using polyline based partition, the accuracy of motion compensation can be improved and the prediction error can be reduced. Curvature analysis of object contour within macroblocks verified the validity of the proposed method. Experimental results show that the proposed method improves the coding efficiency by 0.2-0.3dB in PSNR, when compared to H.264/AVC.

This paper studies the relationship between corporate governance and Merger and Acquisition with 921 Chinese listed companies as our sample. Empirical outcome shows: ownership structure is an important factor which drives a listed company to be MA the proportion of largest shareholdings and the second to the tenth largest shareholding ratio and state share proportion have a significant negative relation with the possibility of being targeted company. However, board of directors' governance has not significant relation with the possibility of being targeted. Whether board chairman and CEO are the same person and independent director's ratio are not closely related with the possibility of being targeted companies.