Liwei WANG Yanduo ZHANG Tao LU Wenhua FANG Yu WANG
Person re-identification (Re-ID) aims to match the same pedestrain identity images across different camera views. Because pedestrians will change clothes frequently for a relatively long time, while many current methods rely heavily on color appearance information or only focus on the person biometric features, these methods make the performance dropped apparently when it is applied to Clohting-Changing. To relieve this dilemma, we proposed a novel Multi Feature Fusion Attention Network (MFFAN), which learns the fine-grained local features. Then we introduced a Clothing Adaptive Attention (CAA) module, which can integrate multiple granularity features to guide model to learn pedestrain's biometric feature. Meanwhile, in order to fully verify the performance of our method on clothing-changing Re-ID problem, we designed a Clothing Generation Network (CGN), which can generate multiple pictures of the same identity wearing different clothes. Finally, experimental results show that our method exceeds the current best method by over 5% and 6% on the VCcloth and PRCC datasets respectively.
Hongzhe LIU Ningwei WANG Xuewei LI Cheng XU Yaze LI
In the neck part of a two-stage object detection network, feature fusion is generally carried out in either a top-down or bottom-up manner. However, two types of imbalance may exist: feature imbalance in the neck of the model and gradient imbalance in the region of interest extraction layer due to the scale changes of objects. The deeper the network is, the more abstract the learned features are, that is to say, more semantic information can be extracted. However, the extracted image background, spatial location, and other resolution information are less. In contrast, the shallow part can learn little semantic information, but a lot of spatial location information. We propose the Both Ends to Centre to Multiple Layers (BEtM) feature fusion method to solve the feature imbalance problem in the neck and a Multi-level Region of Interest Feature Extraction (MRoIE) layer to solve the gradient imbalance problem. In combination with the Region-based Convolutional Neural Network (R-CNN) framework, our Balanced Feature Fusion (BFF) method offers significantly improved network performance compared with the Faster R-CNN architecture. On the MS COCO 2017 dataset, it achieves an average precision (AP) that is 1.9 points and 3.2 points higher than those of the Feature Pyramid Network (FPN) Faster R-CNN framework and the Generic Region of Interest Extractor (GRoIE) framework, respectively.
Hongbin LIN Zheng WU Dong LEI Wei WANG Xiuping PENG
This letter presents a novel tensor voting mechanism — analytic tensor voting (ATV), to get rid of the difficulties in original tensor voting, especially the efficiency. One of the main advantages is its explicit voting formulations, which benefit the completion of tensor voting theory and computational efficiency. Firstly, new decaying function was designed following the basic spirit of decaying function in original tensor voting (OTV). Secondly, analytic stick tensor voting (ASTV) was formulated using the new decaying function. Thirdly, analytic plate and ball tensor voting (APTV, ABTV) were formulated through controllable stick tensor construction and tensorial integration. These make the each voting of tensor can be computed by several non-iterative matrix operations, improving the efficiency of tensor voting remarkably. Experimental results validate the effectiveness of proposed method.
Zhenyu WEI Wei WANG Ben WANG Ping LIU Linshu GONG
Sparse arrays can usually achieve larger array apertures than uniform linear arrays (ULA) with the same number of physical antennas. However, the conventional direction-of-arrival (DOA) estimation algorithms for sparse arrays usually require the spatial smoothing operation to recover the matrix rank which inevitably involves heavy computational complexity and leads to a reduction in the degrees-of-freedom (DOFs). In this paper, a low-complex DOA estimation algorithm by exploiting the discrete Fourier transform (DFT) is proposed. Firstly, the spatial spectrum of the virtual array constructed from the sparse array is established by exploiting the DFT operation. The initial DOA estimation can obtain directly by searching the peaks in the DFT spectrum. However, since the number of array antennas is finite, there exists spectrum power leakage which will cause the performance degradation. To further improve the angle resolution, an iterative process is developed to suppress the spectrum power leakage. Thus, the proposed algorithm does not require the spatial smoothing operation and the computational complexity is reduced effectively. In addition, due to the extention of DOF with the application of the sparse arrays, the proposed algorithm can resolve the underdetermined DOA estimation problems. The superiority of the proposed algorithm is demonstrated by simulation results.
Cho-chin LIN Da-wei WANG Tsan-sheng HSU
We discuss the problem of finding a dominant sequence for sending input data items from a low-end client to a server for computational intensive tasks under the realistic assumption of unpredictable communication behavior. Under this assumption, the client has to send the input data items using a specified sequence to maximize the number of computations performed by the server at any time. The sequence-finding problem is NP-hard for the general case. In this paper, we address three fundamental and useful applications: the product of two polynomials, matrices multiplication and Fast Fourier Transform. We show that the sequence-finding problems of the three applications can be solved optimally in linear time. However, we also show counter examples to rule out any possibility of finding a dominant sequence for sparse cases of the three applications. Finally, a simulation is conducted to show the usefulness of our method.
Ing-Chao LIN Yen-Han LEE Sheng-Wei WANG
Ternary content addressable memory (TCAM), which can store 0, 1, or X in its cells, is widely used to store routing tables in network routers. Negative bias temperature instability (NBTI) and positive bias temperature instability (PBTI), which increase Vth and degrade transistor switching speed, have become major reliability challenges. This study analyzes the signal probability of routing tables. The results show that many cells retain static stress and suffer significant degradation caused by NBTI and PBTI effects. The bit flipping technique is improved and proactive power gating recovery is proposed to mitigate NBTI and PBTI effects. In order to maintain the functionality of TCAM after bit flipping, a novel TCAM cell design is proposed. Simulation results show that compared to the original architecture, the bit flipping technique improves read static noise margin (SNM) for data and mask cells by 16.84% and 29.94%, respectively, and reduces search time degradation by 12.95%. The power gating technique improves read SNM for data and mask cells by 12.31% and 20.92%, respectively, and reduces search time degradation by 17.57%. When both techniques are used, read SNM for data and mask cells is improved by 17.74% and 30.53%, respectively, and search time degradation is reduced by 21.01%.
Haibo YIN Jun-an YANG Wei WANG Hui LIU
Transfer boosting, a branch of instance-based transfer learning, is a commonly adopted transfer learning method. However, currently popular transfer boosting methods focus on binary classification problems even though there are many multi-classification tasks in practice. In this paper, we developed a new algorithm called MultiTransferBoost on the basis of TransferBoost for multi-classification. MultiTransferBoost firstly separated the multi-classification problem into several orthogonal binary classification problems. During each iteration, MultiTransferBoost boosted weighted instances from different source domains while each instance's weight was assigned and updated by evaluating the difficulty of the instance being correctly classified and the “transferability” of the instance's corresponding source domain to the target. The updating process repeated until it reached the predefined training error or iteration number. The weight update factors, which were analyzed and adjusted to minimize the Hamming loss of the output coding, strengthened the connections among the sub binary problems during each iteration. Experimental results demonstrated that MultiTransferBoost had better classification performance and less computational burden than existing instance-based algorithms using the One-Against-One (OAO) strategy.
Jinwei WANG Xirong MA Yuanping ZHU Jizhou SUN
Modern GPUs have evolved to become a more general processor capable of executing scientific and engineering computations. It provides a highly parallel computing environment due to its large number of computing cores, which are suitable for numerous data parallel arithmetic computations, particularly linear algebra operations. The matrix-vector multiplication is one of the most important dense linear algebraic operations. It is applied to a diverse set of applications in many fields and must therefore be fully optimized to achieve a high-performance. In this paper, we proposed a novel auto-tuning method for matrix-vector multiplication on GPUs, where the number of assigned threads that are used to compute one element of the result vector can be auto-tuned according to the size of matrix. On the Nvidia's GPU GTX 650 with the most recent Kepler architecture, we developed an auto-tuner that can automatically select the optimal number of assigned threads for calculation. Based on the auto-tuner's result, we developed a versatile generic matrix-vector multiplication kernel with the CUDA programming model. A series of experiments on different shapes and sizes of matrices were conducted for comparing the performance of our kernel with that of the kernels from CUBLAS 5.0, MAGMA 1.3 and a warp method. The experiments results show that the performance of our matrix-vector multiplication kernel is close to the optimal behavior with increasing of the size of the matrix and has very little dependency on the shape of the matrix, which is a significant improvement compared to the other three kernels that exhibit unstable performance behavior for different shapes of matrices.
Qi LIU Wei WANG Dong LIANG Xianpeng WANG
In this paper, a real-valued reweighted l1 norm minimization method based on data reconstruction in monostatic multiple-input multiple-output (MIMO) radar is proposed. Exploiting the special structure of the received data, and through the received data reconstruction approach and unitary transformation technique, a one-dimensional real-valued received data matrix can be obtained for recovering the sparse signal. Then a weight matrix based on real-valued MUSIC spectrum is designed for reweighting l1 norm minimization to enhance the sparsity of solution. Finally, the DOA can be estimated by finding the non-zero rows in the recovered matrix. Compared with traditional l1 norm-based minimization methods, the proposed method provides better angle estimation performance. Simulation results are presented to verify the effectiveness and advantage of the proposed method.
Wei WANG Xian-peng WANG Yue-hua MA Xin LI
A novel conjugate unitary ESPRIT (CU-ESPRIT) algorithm for the joint direction of departure (DOD), and direction of arrival (DOA), estimation in a bistatic MIMO radar is proposed. A new virtual array is formed by using the properties of noncircular signals, and the properties of the centro-Hermitian matrix are employed to convert the complex-valued data matrix into a real-valued data matrix. Then the real-valued rotational invariance properties of the new virtual array are determined to estimate DODs and DOAs, which are paired automatically. The proposed method provides better angle estimation performance and detects more targets owing to double number of MIMO virtual array elements. Simulation results are presented to verify the effectiveness of the proposed algorithm.
Yingxiao XIANG Chao LI Tong CHEN Yike LI Endong TONG Wenjia NIU Qiong LI Jiqiang LIU Wei WANG
Controlled optimization of phases (COP) is a core implementation in the future intelligent traffic signal system (I-SIG), which has been deployed and tested in countries including the U.S. and China. In such a system design, optimal signal control depends on dynamic traffic situation awareness via connected vehicles. Unfortunately, I-SIG suffers data spoofing from any hacked vehicle; in particular, the spoofing of the last vehicle can break the system and cause severe traffic congestion. Specifically, coordinated attacks on multiple intersections may even bring cascading failure of the road traffic network. To mitigate this security issue, a blockchain-based multi-intersection joint defense mechanism upon COP planning is designed. The major contributions of this paper are the following. 1) A blockchain network constituted by road-side units at multiple intersections, which are originally distributed and decentralized, is proposed to obtain accurate and reliable spoofing detection. 2) COP-oriented smart contract is implemented and utilized to ensure the credibility of spoofing vehicle detection. Thus, an I-SIG can automatically execute a signal planning scheme according to traffic information without spoofing data. Security analysis for the data spoofing attack is carried out to demonstrate the security. Meanwhile, experiments on the simulation platform VISSIM and Hyperledger Fabric show the efficiency and practicality of the blockchain-based defense mechanism.
This paper proposes a new class of Hilbert pairs of almost symmetric orthogonal wavelet bases. For two wavelet bases to form a Hilbert pair, the corresponding scaling lowpass filters are required to satisfy the half-sample delay condition. In this paper, we design simultaneously two scaling lowpass filters with the arbitrarily specified flat group delay responses at ω=0, which satisfy the half-sample delay condition. In addition to specifying the number of vanishing moments, we apply the Remez exchange algorithm to minimize the difference of frequency responses between two scaling lowpass filters, in order to improve the analyticity of complex wavelets. The equiripple behavior of the error function can be obtained through a few iterations. Therefore, the resulting complex wavelets are orthogonal and almost symmetric, and have the improved analyticity. Finally, some examples are presented to demonstrate the effectiveness of the proposed design method.
Zi-wen WANG Guo-rui FENG Ling-yan FAN Jin-wei WANG
The sparse representation models have been widely applied in image super-resolution. The certain optimization problem is supposed and can be solved by the iterative shrinkage algorithm. During iteration, the update of dictionaries and similar patches is necessary to obtain prior knowledge to better solve such ill-conditioned problem as image super-resolution. However, both the processes of iteration and update often spend a lot of time, which will be a bottleneck in practice. To solve it, in this paper, we present the concept of image quality difference based on generalized Gaussian distribution feature which has the same trend with the variation of Peak Signal to Noise Ratio (PSNR), and we update dictionaries or similar patches from the termination strategy according to the adaptive threshold of the image quality difference. Based on this point, we present two sparse representation algorithms for image super-resolution, one achieves the further improvement in image quality and the other decreases running time on the basis of image quality assurance. Experimental results also show that our quantitative results on several test datasets are in line with exceptions.
Yanwei WANG Xiaoqing DING Changsong LIU
This letter has retrained an MQDF classifier on the retraining set, which is constructed by samples locating near classification boundary. The method is evaluated on HCL2000 and HCD Chinese handwriting sets. The results show that the retrained MQDF outperforms MQDF and cascade MQDF on all test sets.
Huaxin XIAO Yu LIU Wei WANG Maojun ZHANG
In consideration of the image noise captured by photoelectric cameras at nighttime, a robust motion detection algorithm based on sparse representation is proposed in this study. A universal dictionary for arbitrary scenes is presented. Realistic and synthetic experiments demonstrate the robustness of the proposed approach.
Lige GE Shengming JIANG Xiaowei WANG Yanli XU Ruoyu FENG Zhichao ZHENG
Along with the fast development of blue economy, wireless communication in oceans has received extensive attention in recent years, and opportunistic networks without any aid from fixed infrastructure or centralized management are expected to play an important role in such highly dynamic environments. Here, link prediction can help nodes to select proper links for data forwarding to reduce transmission failure. The existing prediction schemes are mainly based on analytical models with no adaptability, and consider relatively simple and small terrestrial wireless networks. In this paper, we propose a new link prediction algorithm based on machine learning, which is composed of an extractor of convolutional layers and an estimator of long short-term memory to extract useful representations of time-series data and identify effective long-term dependencies. The experiments manifest that the proposed scheme is more effective and flexible compared with the other link prediction schemes.
Wei WANG Xian-peng WANG Xin LI
A low-complexity method for angle estimation in Multiple-input multiple-output radar (MIMO) radar is presented. In this approach, the signal subspace can be spanned by the orthogonal vectors which are obtained by Multi-stage Wiener Filter (MSWF), then the ESPRIT method can be used to estimate direction of departures (DODs) and direction of arrivals (DOAs). Compared with the conventional ESPRIT algorithm, the proposed method does not involve estimation of the covariance matrix and its eigen-decomposition, which alleviates remarkably the computational complexity. Moreover, the proposed method achieves the similar angle estimation performance. Simulation results are presented to verify the efficiency of the proposed method.
Guiping JIN Guangde ZENG Long LI Wei WANG Yuehui CUI
A triple-band CP rectenna for ambient RF energy harvesting is presented in this paper. A simple broadband CP slot antenna has been proposed with the bandwidth of 51.1% operating from 1.53 to 2.58GHz, which can cover GSM-1800, UMTS-2100 and 2.45GHz WLAN bands. Accordingly, a triple-band rectifying circuit is designed to convert RF energy in the above bands, with the maximum RF-DC conversion efficiency of 42.5% at a relatively low input power of -5dBm. Additionally, the rectenna achieves the maximum conversion efficiency of 12.7% in the laboratory measurements. The measured results show a good performance in the laboratory measurements.
Junqi ZHANG Lina NI Jing YAO Wei WANG Zheng TANG
Kennedy has proposed the bare bones particle swarm (BBPS) by the elimination of the velocity formula and its replacement by the Gaussian sampling strategy without parameter tuning. However, a delicate balance between exploitation and exploration is the key to the success of an optimizer. This paper firstly analyzes the sampling distribution in BBPS, based on which we propose an adaptive BBPS inspired by the cloud model (ACM-BBPS). The cloud model adaptively produces a different standard deviation of the Gaussian sampling for each particle according to the evolutionary state in the swarm, which provides an adaptive balance between exploitation and exploration on different objective functions. Meanwhile, the diversity of the swarms is further enhanced by the randomness of the cloud model itself. Experimental results show that the proposed ACM-BBPS achieves faster convergence speed and more accurate solutions than five other contenders on twenty-five unimodal, basic multimodal, extended multimodal and hybrid composition benchmark functions. The diversity enhancement by the randomness in the cloud model itself is also illustrated.
Yuanyuan XU Wei LI Wei WANG Dan WU Lai HE Jintao HU
A 19.1-to-20.4 GHz sigma-delta fractional-N frequency synthesizer with two-point modulation (TPM) for frequency modulated continuous wave (FMCW) radar applications is presented. The FMCW synthesizer proposes a digital and voltage controlled oscillator (D/VCO) with large continuous frequency tuning range and small digital controlled oscillator (DCO) gain variation to support TPM. By using TPM technique, it avoids the correlation between loop bandwidth and chirp slope, which is beneficial to fast chirp, phase noise and linearity. The start frequency, bandwidth and slope of the FMCW signal are all reconfigurable independently. The FMCW synthesizer achieves a measured phase noise of -93.32 dBc/Hz at 1MHz offset from a 19.25 GHz carrier and less than 10 µs locking time. The root-mean-square (RMS) frequency error is only 112 kHz with 94 kHz/µs chirp slope, and 761 kHz with a fast slope of 9.725 MHz/µs respectively. Implemented in 65 nm CMOS process, the synthesizer consumes 74.3 mW with output buffer.