Segmentation algorithm of complex ore images based on templates transformation and reconstruction

Lots of noises and heterogeneous objects with various sizes coexist in a complex image,such as an ore image;the classical image thresholding method cannot effectively distinguish between ores.To segment ore objects with various sizes simultaneously,two adaptive windows in the image were chosen for each pixel;the gray value of windows was calculated by Otsu＇s threshold method.To extract the object skeleton,the definition principle of distance transformation templates was proposed.The ores linked together in a binary image were separated by distance transformation and gray reconstruction.The seed region of each object was picked up from the local maximum gray region of the reconstruction image.Starting from these seed regions,the watershed method was used to segment ore object effectively.The proposed algorithm marks and segments most objects from complex images precisely.

The Line Clipping Algorithm Basing on Affine Transformation

A new algorithm for clipping line segments by a rectangular window on rectangular coordinate system is presented in this paper. The algorithm is very different to the other line clipping algorithms. For the line segments that cannot be identified as completely inside or outside the window by simple testings, this algorithm applies affine transformations (the shearing transformations) to the line segments and the window, and changes the slopes of the line segments and the shape of the window. Thus, it is clear for the line segment to be outside or inside of the window. If the line segments intersect the window, the algorithm immediately (no solving equations) gets the intersection points. Having applied the inverse transformations to the intersection points, the algorithm has the final results. The algorithm is successful to avoid the complex classifications and computations. Besides, the algorithm is effective to simplify the processes of finding the intersection points. Comparing to some classical algorithms, the algorithm of this paper is faster for clipping line segments and more efficient for calculations.

Numerical Experiments and Structural Transformation of Non-Linear Evolution in Ill-posed Systems

Numerical experiments on non-linear equations of the 1st-and 3rd-order derivatives have been carried out through structural analyses in the phase space according to the numerical instability of ill-posed systems,with changes of initial values and parameters,etc..The results show that the quantitative instability in an ill-posed system may reveal reversed transformation in system evolution by structural representation,and confirm A·Dauglas' theorem that "a non-linear equation does not satisfy the existence of the initial value in a linear well-posed system".

GENERALIZED CANONICAL TRANSFORMATION AND SYMPLECTIC ALGORITHM OF THE AUTONOMOUS BIRKHOFFIAN SYSTEMS

The Birkhoff systems are the generalization of the Hamiltonian systems. Generalized canonical transformations are studied. The symplectic algorithm of the Hamiltonian systems is extended into that of the Birkhofflan systems. Symplectic differential scheme of autonomous Birkhoffian systems vas structured and discussed by introducing the Kailey Transformation.

Design of Sharp 2D Multiplier-Less Circularly Symmetric FIR Filter Using Harmony Search Algorithm and Frequency Transformation

In this paper, we present a novel and efficient method for the design of a sharp, two dimensional (2D) wideband, circularly symmetric, FIR filter. First of all, a sharp one dimensional (1D) infinite precision FIR filter is designed using the Frequency Response Masking (FRM) technique. This filter is converted into a multiplier-less filter by representing it in the Canonic Signed Digit (CSD) space. The design of the FRM filter in the CSD space calls for the use of a discrete optimization technique. To this end, a new optimization approach is proposed using a modified Harmony Search Algorithm (HSA). HSA is modified in such a way that, in every exploitation and exploration phase, the candidate solutions turns out to be integers. The 1D FRM multiplier-less filter, is in turn transformed to the 2D equivalent using the recently proposed multiplier-less transformations namely, T1 and T2. These transformations are successful in generating circular contours even for wideband filters. Since multipliers are the most power consuming elements in a 2D filter, the multiplier-less realization calls for reduced power consumption as well as computation time. Significant reduction in the computational complexity and computation time are the highlights of our proposed design technique. Besides, the proposed discrete optimization using modified HSA can be used to solve optimization problems in other engineering disciplines, where the search space consists of integers.

The ridgelet transform with non-linear threshold for seismic noise attenuation in marine carbonates

Wavelet transforms have been successfully used in seismic data processing with their ability for local time - frequency analysis. However, identification of directionality is limited because wavelet transform coefficients reveal only three spatial orientations. Whereas the ridgelet transform has a superior capability for direction detection and the ability to process signals with linearly changing characteristics. In this paper, we present the issue of low signal-to-noise ratio （SNR） seismic data processing based on the ridgelet transform. Actual seismic data with low SNR from south China has been processed using ridgelet transforms to improve the SNR and the continuity of seismic events. The results show that the ridgelet transform is better than the wavelet transform for these tasks.

基于RoBERTa和图增强Transformer的序列推荐方法

自推荐系统出现以来,有限的数据信息就一直制约着推荐算法的进一步发展。为降低数据稀疏性的影响,增强非评分数据的利用率,基于神经网络的文本推荐模型相继被提出,但主流的卷积或循环神经网络在文本语义理解和长距离关系捕捉方面存在明显劣势。为了更好地挖掘用户与商品之间的深层潜在特征,进一步提高推荐质量,提出一种基于Ro BERTa和图增强Transformer的序列推荐(RGT)模型。引入评论文本数据,首先利用预训练的Ro BERTa模型捕获评论文本中的字词语义特征,初步建模用户的个性化兴趣,然后根据用户与商品的历史交互信息,构建具有时序特性的商品关联图注意力机制网络模型,通过图增强Transformer的方法将图模型学习到的各个商品的特征表示以序列的形式输入Transformer编码层,最后将得到的输出向量与之前捕获的语义表征以及计算得到的商品关联图的全图表征输入全连接层,以捕获用户全局的兴趣偏好,实现用户对商品的预测评分。在3组真实亚马逊公开数据集上的实验结果表明,与Deep FM、Conv MF等经典文本推荐模型相比,RGT模型在均方根误差(RMSE)和平均绝对误差(MAE)2种指标上有显著提升,相较于最优对比模型最高分别提升4.7%和5.3%。

引入Transformer的道路小目标检测

针对道路场景中检测小目标时漏检率较高、检测精度低的问题,提出一种引入Transformer的道路小目标检测算法。在原YOLOv4算法基础上,对多尺度检测进行改进,把浅层特征信息充分利用起来;设计ICvT(improved convolutional vision transformer)模块捕获特征内部的相关性,获得上下文信息,提取更加全面丰富的特征;在网络特征融合部分嵌入改进后的空间金字塔池化模块,在保持较小计算量的同时增加特征图的感受野。实验结果表明,在KITTI数据集上,算法检测精度达到91.97%,与YOLOv4算法相比,mAP提高了2.53%,降低了小目标的漏检率。

Anti-aliasing nonstationary signals detecion algorithm based on interpolation in the frequency domain using the short time Fourier transform

To eliminate the aliasing that appeared during the measurement of multi-components nonstationary signals, a novel kind of anti-aliasing algorithm based on the short time Fourier transform （STFT） is brought forward. First the physical essence of aliasing that occurs is analyzed; second the interpolation algorithm model is setup based on the Hamming window; then the fast implementation of the algorithm using the Newton iteration method is given. Using the numerical simulation the feasibility of algorithm is validated. Finally, the electrical circuit experiment shows the practicality of the algorithm in the electrical engineering.

基于INGO-Transformer的模拟电路元件故障预测

针对模拟电路元件易受外部环境影响发生故障、故障特征提取困难、无法准确预测及诊断元件故障等问题,基于Transformer模型提出改进INGO-Transformer方法。采用小波包分解(WPD)对原始数据进行特征提取,使用特征向量之间的三角距离来表征模拟电路中元件的退化状态,使用INGO优化Transformer的训练超参数构建预测模型。以Sallen-Key带通滤波电路与镜像电流源电路为预测实验对象进行故障预测实验,采用MAE与MSE作为故障预测模型评价指标,两组实验电路10次实验平均MAE、MSE结果分别为4.2162e-04、4.1906e-07和0.0017、1.9625e-05。仿真结果表明,所提方法在模拟电路单一元件故障预测中具有较高的准确性与较强的泛化能力。

Spectral matching algorithm based on nonsubsampled contourlet transform and scale-invariant feature transform

A new spectral matching algorithm is proposed by us- ing nonsubsampled contourlet transform and scale-invariant fea- ture transform. The nonsubsampled contourlet transform is used to decompose an image into a low frequency image and several high frequency images, and the scale-invariant feature transform is employed to extract feature points from the low frequency im- age. A proximity matrix is constructed for the feature points of two related images. By singular value decomposition of the proximity matrix, a matching matrix （or matching result） reflecting the match- ing degree among feature points is obtained. Experimental results indicate that the proposed algorithm can reduce time complexity and possess a higher accuracy.

Lane Recognition Algorithm Using the Hough Transform Based on Complicated Conditions

At present, most lane line detection methods are aimed at simple road surface. There is still no good solution for the situation that the lane line contains arrow, text and other signs. The edge left by markers such as arrow and text will interfere with the detection of lane lines. In view of the situation of arrow mark and text mark interference between lane lines, the paper proposes a new processing algorithm. The algorithm consists of four parts, Gaussian blur, image graying processing, DLD-threshold (Dark-Light-Dark-threshold) algorithm, correlation filter edge extraction and Hough transform. Among them, the DLD-threshold algorithm and related filters are mainly used to remove the identification interference between lane lines. The test results on the Caltech Lanes dataset are given at the end of the article. The result of verification of this algorithm showed a max recognition rate of 97.2%.

A-E equation of potential field transformations in the wavenumber domain and its application

A shift sampling theory established by author （1997a） is a generalization of Fourier transform computation theory. Based on this theory, I develop an Algorithm-Error （A-E） equation of potential field transformations in the wavenumber domain, which not only gives a more flexible algorithm of potential field transformations, but also reveals the law of error of potential field transformations in the wavenumber domain. The DFT0η η（0.5, 0.5） reduction-to-pole （RTP） technique derived from the A-E equation significantly improves the resolution and accuracy of RTP anomalies at low magnetic latitudes, including the magnetic equator. The law （origin, form mechanism, and essential properties） of the edge oscillation revealed by the A-E equation points out theoretically a way of improving the effect of existing padding methods in high-pass transformations in the wavenumber domain.

基于IWOA-Transformer的磨煤机故障预警

提出了一种基于改进鲸鱼算法优化Transformer网络超参数(IWOA-Transformer)的故障预警方法。该方法利用非线性收敛系数和高斯变异对鲸鱼算法(WOA)进行改进,以提高WOA的收敛速度和避免其陷入局部最优;再采用改进鲸鱼算法(IWOA)优化Transformer的超参数,建立磨煤机故障预警模型;然后,通过预测值和实际值的相似度函数确定自适应阈值,结合专家系统判断故障类型并提出解决方案,实现磨煤机故障预警;最后,以某350 MW热电机组中速磨煤机为例进行故障预警试验。结果表明:所提IWOA-Transformer模型可显著提高预警速度和准确率,具有工程实用价值。

An Algorithm for Ship Wake Detection from the SAR Images Using the Radon Transform and Morphological Image Processing

Using the Radon transform and morphological image processing, an algorithm for ship's wake detection in the SAR (synthetic aperture radar) image is developed. Being manipulated in the Radon space to invert the gray-level and binary images, the linear texture of ship wake in oceanic clutter can be well detected. It has been applied to the automatic detection of a moving ship from the SEASAT SAR image. The results show that this algorithm is well robust in a strong noisy background and is not very sensitive to the threshold parameter and the working window size.

RESEARCH OF PROBLEMS ON REALIZING DIRECT ALGORITHM OF WAVELET TRANSFORM

Direct algorithm of wavelet transform (WT) is the numerical algorithmobtained from the integral formula of WT by directly digitization. Some problems on realizing thealgorithm are studied. Some conclusions on the direct algorithm of discrete wavelet transform (DWT),such as discrete convolution operation formula of wavelet coefficients and wavelet components,sampling principle and technology to wavelets, deciding method for scale range of wavelets, measuresto solve edge effect problem, etc, are obtained. The realization of direct algorithm of continuouswavelet transform (CWT) is also studied. The computing cost of direct algorithm and Mallat algorithmof DWT are still studied, and the computing formulae are obtained. These works are beneficial todeeply understand WT and Mallat algorithm. Examples in the end show that direct algorithm can alsobe applied widely.

The Parameter Determination in the Design of 3-D Spherically Symmetric FIR Filters via the McClellan Transformation

In the design of 3-D spherically symmetric FIR filters via the McClellan transformation, two methods are proposed to determine the transformation parameters. The first is to improve the original 3-D algorithm by exploiting the 2-D effective methods in 3-D. This method can change the constrained optimization algorithm into the unconstrained one and makes the design easier to realize. The second method is to solve the coupled equations under constrained conditions and a set of ideal parameters can be gotten. The design example shows that the two methods are all efficient and easier than the original algorithm.

Three-dimensional transformation of magnetization direction and magnetic field component at low latitudes based on vertical relationship

The transformation of the magnetization direction and the magnetic fi eld component is one of the important methods in magnetic data processing and transformation,which can be conducted in both wavenumber and spatial domains.The transformation method in the wavenumber domain has simpler processing expression and higher processing effi ciency than in the spatial domain;however,they are unstable at low latitude.In this paper,the conclusion that the sum is 0 of two vertical magnetic fi eld components(magnetization inclinations are also perpendicular)in 2D is used for the 3D transformation of the magnetization direction and the magnetic field component.In addition,the transformation method at low latitudes based on vertical relationship(VMT)is proposed,which is an iterative algorithm that converts the transformation of the magnetization direction and the magnetic field component at the low latitude into the high latitude.This method restrains the instability of transformation of constant and variable magnetization direction and magnetic fi eld components in low latitudes.The accuracy,stability,and practicality are verifi ed from synthetic models and real data.

Power Quality Disturbance Classification Method Based on Wavelet Transform and SVM Multi-class Algorithms

The accurate identification and classification of various power quality disturbances are keys to ensuring high-quality electrical energy. In this study, the statistical characteristics of the disturbance signal of wavelet transform coefficients and wavelet transform energy distribution constitute feature vectors. These vectors are then trained and tested using SVM multi-class algorithms. Experimental results demonstrate that the SVM multi-class algorithms, which use the Gaussian radial basis function, exponential radial basis function, and hyperbolic tangent function as basis functions, are suitable methods for power quality disturbance classification.

New SAR Imaging Algorithm via the Optimal Time-Frequency Transform Domain

To address the low-resolution imaging problem in relation to traditional Range Doppler(RD)algorithm,this paper intends to propose a new algorithm based on Fractional Fourier Transform(FrFT),which proves highly advantageous in the acquisition of high-resolution Synthetic Aperture Radar(SAR)images.The expression of the optimal order of SAR range signals using FrFT is deduced in detail,and the corresponding expression of the azimuth signal is also given.Theoretical analysis shows that,the optimal order in range(azimuth)direction,which turns out to be very unique,depends on the known imaging parameters of SAR,therefore the engineering practicability of FrFT-RD algorithm can be greatly improved without the need of order iteration.The FrFT-RD algorithm is established after an analysis of the optimal time-frequency transform.Experimental results demonstrate that,compared with traditional RD algorithm,the main-lobe width of the peak-point target of FrFT-RD algorithm is narrow in both range and azimuth directions.While the peak amplitude of the first side-lobe is reduced significantly,those of other side-lobes also drop in various degrees.In this way,the imaging resolution of range and azimuth can be increased considerably.