基于卷积神经网络的包号识别方法及其应用

以YOLOv3神经网络为基础,提出一种炼钢车间内包号识别方法。首先收集了现场包号的图像组成原始数据集,之后对其进行筛选、标注、图像预处理等形成训练数据集。接着,借助K-means聚类运算得到网络训练所需的锚框尺寸,并针对包号图像的特点调整了网络模型参数。最后,在训练数据集上训练网络得到识别模型。识别模型的输出经后处理,应用在天车物流跟踪系统中。结果显示,该包号识别方法的准确率在98%以上。此外,在生产环境长期运行的实践表明,本识别方法可满足天车物流跟踪系统对包号识别准确性、稳定性和实时性的要求。

3D包号莫忘“止损” 必要时候应该做到当断则断

在买3D时，不少彩民都会选择包号法，对包号法也有一定的心得，但需要注意的是包号还须莫忘“止损”。

监狱网格化包号责任制的实践探索——基于福建监狱工作

实行网格化包号责任制是推动监狱由"监管看守型"向"教育改造型"升级的有力举措。福建监狱积极推进网格化包号新举措,将号房作为管理和改造的基本单元,对罪犯实现全程动态监管和精细化改造。此举在维护监狱安全稳定、化解服刑人员矛盾、实现监狱教育改造工作提质增效等方面取得了实效,但是在运行实践中也发现一些问题。从实践出发,可从工作职责分工、考评激励机制、标准化规范化操作、网格间信息联通共享等方面深化推进网格化包号责任制。

Application of wavelet package filtering in the de-noising of fiber optic gyroscopes

To reduce the drift error existing in the output signal of fiber optic gyroscopes （FOG）, a mathematical model of the FOG output signal is set up; the error characteristics of the FOG output signal are analyzed, and semi-soft threshold filtering is chosen based on the comparison of hard threshold and soft threshold filtering. The semi-soft threshold wavelet package filtering method is applied in the filtering of the FOG output signal. Experiments of the stationary and dynamic FOG output signals filtered with the wavelet package analysis are carried out in a lab environment, respectively. Experiments done with the real-time measured FOG signal show that the method of semi-soft threshold wavelet package filtering reduces the mean square error from 5 （°）/h to 1 （°）/h, so it is effective in eliminating the white noises and the fractal noises existing in the FOG. The novel method proposed here is proved valid in reducing the FOG drift error, satisfying the technical demands of high precision and realtime processing.

Classification of forearm action surface EMG signals based on fractal dimension

Surface electromyogram （EMG） signals were identified by fractal dimension.Two patterns of surface EMG signals were acquired from 30 healthy volunteers＇ right forearm flexor respectively in the process of forearm supination （FS） and forearm pronation （FP）.After the raw action surface EMG （ASEMG） signal was decomposed into several sub-signals with wavelet packet transform （WPT）,five fractal dimensions were respectively calculated from the raw signal and four sub-signals by the method based on fuzzy self-similarity.The results show that calculated from the sub-signal in the band 0 to 125 Hz,the fractal dimensions of FS ASEMG signals and FP ASEMG signals distributed in two different regions,and its error rate based on Bayes decision was no more than 2.26%.Therefore,the fractal dimension is an appropriate feature by which an FS ASEMG signal is distinguished from an FP ASEMG signal.

净毛率准确性浅析

对于开放的羊毛市场来说,“公平交易”是购销双方的共同愿望。作为结价依据的“净毛公量”是交易双方的生命线,净毛率测试是公量的重要组成部分。GB/T1523-93标准要求公量误差控制在5％以内,严格执行“公量误差”则是我们纤检试验人员义不容辞的责任。我们在总结多年净毛率测试的基础上。

Quantity of Fertilizer for a New Late-maturing Heading Mustard(Brassica juncea) Cultivar Yongbao 2 at the Heading Stage

[Objective] The study was to determine the optimal quantity of fertilizer required for Yongbao 2 during heading. [Method] According to a randomized block design with three replicates, four fertilizer treatments and one blank control were set in field trials. Each plot had a net area of 66.7 m2, Finally, the mustard yield and economic benefit per plot among the treatments were compared. [Result] There was no significant difference in yield among the four fertilizer treatments. The agronomic traits and yield of Yongbao 2 were improved when the quantity of fertilizer was gradually increased. The treatment with 750 kg/hm2 compound fertilizer during mus- tard heading had the highest yield. However, the treatment with 450 kg/hm2 of compound fertilizer during mustard heading had the highest economic benefit. By considering both yield and benefit, 450 kg/hm2 was the optimal quantity of fertilizer required for Yongbao 2 during heading. [Conclusion] The results will provide references for future application and extension of Yongbao 2.

Performance optimization for multicast packet authentication

In secure multicast, one of the challenging problems is the authentication of multicast packets. This paper presents a novel scheme to address this problem, which combines ideas in both the hash tree schemes and the hash chain schemes. In this scheme, a group of packets is partitioned into equal-sized subgroups. Then a Merkle hash tree is built for each subgroup of packets, and the hash value of every root is appended to preceding packets to form hash chains. Its performance is analyzed and simulated using Biased Coin Toss loss model and 2-state Markov Chain loss model, respectively. Compared with the original hash chain schemes, results show that this scheme is much more efficient in term of communication overhead.

A Method of Peak Detecting for Nanosecond Pulse in Optical Performance Monitoring System

In optical performance monitoring system,the analog to digital converter is needed to detect the peak of nanosecond pulse and get the signal envelope.A scheme based on a designed anti-aliasing filter and analog to digital converter is proposed to broaden the nanosecond pulse and make it easier for the analog to digital converter to catch the peak of the nanosecond pulse.The experimental results demonstrate that,with the proposed scheme,the optical performance system needs less time to get the recovered eye-diagram of high speed optical data signal,and is robust to phase mismatch in the analog to digital converter circuit.

Quality improvement of the dual-wavelength signals in DLOB via power equalization

The data packet buffeting can be implemented in dual loop optical buffer （DLOB） through cross-phase modulation （XPM） between the control signal and the data packets. However, the output equality of dual wavelength signals will degrade due to the unbalanced gain and phase shift. This problem can be resolved by power oqualizafion using the saturation characteristics of SOA. The data packets with 2.5 Gb/s can be buffered in DLOB for 16 cycles corresponding to 16.5 μs.The extinction ratio of the output packet is 8 dB while the S/N is 8.1 dB.

Motion Classification of EMG Signals Based on Wavelet Packet Transform and LS-SVMs Ensemble

This paper presents an effective method for motion classification using the surface electromyographic (sEMG) signal collected from the forearm. Given the nonlinear and time-varying nature of EMG signal, the wavelet packet transform (WPT) is introduced to extract time-frequency joint information. Then the multi-class classifier based on the least squares support vector machine (LS-SVM) is constructed and verified in the various motion classification tasks. The results of contrastive experiments show that different motions can be identified with high accuracy by the presented method. Furthermore, compared with other classifiers with different features, the performance indicates the potential of the SVM techniques combined with WPT in motion classification.

A Symbol Timing Recovery Algorithm of M-PSK Signals for Burst Modem Applications with Small Packet Size

Feedforward symbol timing recovery techniques are particularly important for initial acquisition in burst modems. However, these techniques either have large calculation burden or sensitive to frequency offsets. In this paper, we proposed an efficient symbol timing recovery algorithm of MPSK signals named OMQ(Ordered Maximum power using Quadratic approximation partially) algorithm which is based on the Quadratic Approximation(QA) algorithm. We used ordered statistic sorting method to reduce the computational complexity further, meanwhile maximum mean power principle was used to decrease frequency offset sensitivity. The proposed algorithm adopts estimation-down sampling structure which is suitable for small packet size transmission. The results show that, while comparing with the QA algorithm, the computational complexity is reduced by 75% at most when 8 samples per symbol are used. The proposed algorithm shows better performance in terms of the jitter variance and sensitivity to frequency offsets.

Application of Wavelet Packet Energy Spectrum to Extract the Feature of the Pulse Signal

The wavelet packet is presented as a new kind of multiscale analysis technique followed by Wavelet analysis. The fundamental and realization arithmetic of the wavelet packet analysis method are described in this paper. A new application approach of the wavelet packed method to extract the feature of the pulse signal from energy distributing angle is expatiated. It is convenient for the microchip to process and judge by using the wavelet packet analysis method to make the pulse signals quantized and analyzed. Kinds of experiments are simulated in the lab, and the experiments prove that it is a convenient and accurate method to extract the feature of the pulse signal based on wavelet packed-energy spectrumanalysis.

A method to compress vibration signals using wavelet packet transformation combined with sub-band vector quantization

A novel compression method for mechanical vibrating signals,binding with sub-band vector quantization(SVQ) by wavelet packet transformation(WPT) and discrete cosine transformation(DCT) is proposed.Firstly,the vibrating signal is decomposed into sub-bands by WPT.Then DCT and adaptive bit allocation are done per sub-band and SVQ is performed in each sub-band.It is noted that,after DCT,we only need to code the first components whose numbers are determined by the bits allocated to that sub-band.Through an actual signal,our algorithm is proven to improve the signal-to-noise ratio(SNR) of the reconstructed signal effectively,especially in the situation of lowrate transmission.

Software Package Evaluation for Lyapunov Exponent and Others Features of Signals Evaluating Condition Monitoring Performance of Nonlinear Dynamic Systems

Efficient use of industrial equipment, increase its availability, safety and economic issues spur strong research on maintenance programs based on their operating conditions. Machines normally operate in a linear range, but when malfunctions occur, nonlinear behavior might set in. By studying and comparing five nonlinear features, which listed in decreasing order by their damage detection capability are： LLE （largest Lyapunov exponent）, embedded dimension, Kappa determinism, time delay and cross error values; i.e., LLE performs best. Using somewhat similar ideas from Chaos control, i.e., vary the ＂mass imbalance＂ forcing parameters, we aim to stabilize the Lorenz equation. Quite interestingly, for certain imbalance excitation values, the system is stabilized. The previous even when paradigmatically chaotic parameters for Lorenz system are used （plus our forcing terms）. This quasi-control approach is validated studying signals obtained from the previously mentioned lab test. Finally, it is concluded that analyzing and comparing nonlinear features extracted from baseline vs. malfunction condition （test acquired）, one might increase the efficiency and the performance of machine condition monitoring.

Joint Estimation of Carrier Frequency and Phase Offset Based on Pilot Symbols in Quasi-Constant Envelope OFDM Satellite Systems

Spectral efficiency and energy efficiency are two important performance indicators of satellite systems. The Quasi-Constant Envelope Orthogonal Frequency Division Multiplexing(QCE-OFDM) technique can achieve both high spectral efficiency and low peak-to-average power ratio(PAPR). Therefore, the QCE-OFDM technique is considered as a promising candidate multi-carrier technique for satellite systems. However, the Doppler effect will cause the carrier frequency offset(CFO), and the non-ideal oscillator will cause the carrier phase offset(CPO) in satellite systems. The CFO and CPO will further result in the bit-error-rate(BER) performance degradation. Hence, it is important to estimate and compensate the CFO and CPO. This paper analyzes the effects of both CFO and CPO in QCE-OFDM satellite systems. Furthermore, we propose a joint CFO and CPO estimation method based on the pilot symbols in the frequency domain. In addition, the optimal pilot symbol structure with different pilot overheads is designed according to the minimum Cramer-Rao bound(CRB) criterion. Simulation results show that the estimation accuracy of the proposed method is close to the CRB.

Photonic envelope detection and fiber transmission of 24 GHz IR-UWB signal based on phase modulation

A novel scheme for photonic envelope detection and fiber transmission of 24 GHz impulse radio ultra-wideband(IRUWB) signal is proposed based on phase modulator(PM). In the system, an optics assisted envelope detection unit(OAEDU) is used for filtering one of the first sidebands at the output of PM, then this narrow band optical signal transfers over single-mode fiber(SMF), and the envelope of 24 GHz IR-UWB signal is obtained after photodetection(PD) and low pass filter(LPF). The numerical simulation results show that the combination of PM and OAEDU can alleviate the fiber chromatic dispersion(CD) effectively. The proposed system may provide a simple and cost-effective solution for IR-UWB receiver.