By mcans of stable attractors of discret Hopfield neural network (DHNN) , anew class of nonlinear error control codes is sugsested and some relativetheorems are presented. A kind of single error control codes is also ...By mcans of stable attractors of discret Hopfield neural network (DHNN) , anew class of nonlinear error control codes is sugsested and some relativetheorems are presented. A kind of single error control codes is also given forillustrating this new approach.展开更多
Presents the fuzzy neural network optimized by hybrid coded genetic algorithm of decimal encoding and binary encoding, the searching ability and stability of genetic algorithms enhanced by using binary encoding during...Presents the fuzzy neural network optimized by hybrid coded genetic algorithm of decimal encoding and binary encoding, the searching ability and stability of genetic algorithms enhanced by using binary encoding during the crossover operation and decimal encoding during the mutation operation, and the way of accepting new individuals by probability adopted, by which a new individual is accepted and its parent is discarded when its fitness is higher than that of its parent, and a new individual is accepted by probability when its fitness is lower than that of its parent. And concludes with calculations made with an example that these improvements enhance the speed of genetic algorithms to optimize the fuzzy neural network controller.展开更多
A method of assigning binary indexes to codevectors in vector quantization (VQ)system, which is called pseudo-Gray coding, is presented in this paper by constructing a kind of Hopfield neural network. Pseudo-Gray codi...A method of assigning binary indexes to codevectors in vector quantization (VQ)system, which is called pseudo-Gray coding, is presented in this paper by constructing a kind of Hopfield neural network. Pseudo-Gray coding belongs to joint source/channel coding, which could provide a redundancy-free error protection scheme for VQ of analog signals when the binary indexes of signal codevectors are used as channel symbols on a discrete memoryless channel. Since pseudo-Gray coding is of combinatorial optimization problems which are NP-complete problems,globally optimal solutions are generally impossible. Thus, a kind of Hopfield neural network is used by constructing suitable energy function to get sub-optimal solutions. This kind of Hop field neural network is easily modified to solve simplified version of pseudo-Gray coding for single bit-error channel model. Simulating experimental results show that the method introduced here could offer good performances.展开更多
A new 3D surface contouring and ranging system based on digital fringe projection and phase shifting technique is presented. Using the phase-shift technique, points cloud with high spatial resolution and limited accur...A new 3D surface contouring and ranging system based on digital fringe projection and phase shifting technique is presented. Using the phase-shift technique, points cloud with high spatial resolution and limited accuracy can be generated. Stereo-pair images obtained from two cameras can be used to compute 3D world coordinates of a point using traditional active triangulation approach, yet the camera calibration is crucial. Neural network is a well-known approach to approximate a nonlinear system without an explicit physical model, in this work it is used to train the stereo vision application system to calculating 3D world coordinates such that the camera calibration can be bypassed. The training set for neural network consists of a variety of stereo-pair images and the corresponding 3D world coordinates. The picture elements correspondence problem is solved by using projected color-coded fringes with different orientations. Color imbalance is completely eliminated by the new color-coded method. Once the high accuracy correspondence of 2D images with 3D points is acquired, high precision 3D points cloud can be recognized by the well trained net. The obvious advantage of this approach is that high spatial resolution can be obtained by the phase-shifting technique and high accuracy 3D object point coordinates are achieved by the well trained net which is independent of the camera model works for any type of camera. Some experiments verified the performance of the method.展开更多
针对神经网络超参数优化效果差、容易陷入次优解和优化效率低的问题,提出一种基于改进实数编码遗传算法(IRCGA)的深度神经网络超参数优化算法——IRCGA-DNN(IRCGA for Deep Neural Network)。首先,采用实数编码方式表示超参数的取值,使...针对神经网络超参数优化效果差、容易陷入次优解和优化效率低的问题,提出一种基于改进实数编码遗传算法(IRCGA)的深度神经网络超参数优化算法——IRCGA-DNN(IRCGA for Deep Neural Network)。首先,采用实数编码方式表示超参数的取值,使超参数的搜索空间更灵活;然后,引入分层比例选择算子增加解集多样性;最后,分别设计了改进的单点交叉和变异算子,以更全面地探索超参数空间,提高优化算法的效率和质量。基于两个仿真数据集,验证IRCGA-DNN的毁伤效果预测性能和收敛效率。实验结果表明,在两个数据集上,与GA-DNN(Genetic Algorithm for Deep Neural Network)相比,所提算法的收敛迭代次数分别减少了8.7%和13.6%,均方误差(MSE)相差不大;与IGA-DNN(Improved GA-DNN)相比,IRCGA-DNN的收敛迭代次数分别减少了22.2%和13.6%。实验结果表明,所提算法收敛速度和预测性能均更优,能有效处理神经网络超参数优化问题。展开更多
相比于高效视频编码(high efficiency video coding,HEVC)标准,新一代编码标准多功能视频编码(versatile video coding,VVC)引入了很多新的技术,其中包括四叉树(quadtree,QT)和多类型树(multi-type tree,MTT)划分,MTT划分由HEVC中的QT...相比于高效视频编码(high efficiency video coding,HEVC)标准,新一代编码标准多功能视频编码(versatile video coding,VVC)引入了很多新的技术,其中包括四叉树(quadtree,QT)和多类型树(multi-type tree,MTT)划分,MTT划分由HEVC中的QT划分延伸而来。新划分方法提高了压缩效率,但导致编码时间急剧增加。为了降低编码复杂度,提出了一种结合深度学习方法和MTT方向早期判决的快速帧内编码算法。首先使用轻量级的卷积神经网络(convolutional neural network,CNN)对QT和部分MTT进行预测划分,其余MTT则采用提前预测MTT划分方向的方法作进一步的优化。实验结果表明,所提方法能够大幅降低编码复杂度,相比于原始编码器的编码时间减少了74.3%,且只有3.3%的码率损失,性能优于对比的方法。展开更多
为了提升高效视频编码(High Efficiency Video Coding,HEVC)帧内编码的实时性能,本文提出的方法利用了引入偶数边长与步长的卷积核以及自注意力机制的轻量级卷积网络来预测编码树单元(Coding Tree Unit,CTU)的帧内划分结构,从而减少了...为了提升高效视频编码(High Efficiency Video Coding,HEVC)帧内编码的实时性能,本文提出的方法利用了引入偶数边长与步长的卷积核以及自注意力机制的轻量级卷积网络来预测编码树单元(Coding Tree Unit,CTU)的帧内划分结构,从而减少了编码器对CTU进行四叉树递归遍历划分的编码时间。原始编码策略中粗模式决策通过基于残差经哈德曼变换的预测残差绝对值总和(Sum of Absolute Transformed Difference,SATD)的损失值来估计率失真优化过程中的率失真损失值来进行加速,但仍会耗费一定的编码时间。提出一种方法通过采样搜索的方式减少粗模式决策过程中计算的模式数,从35种模式降低到了18种模式,降低了粗模式决策过程中计算估计损失值的时间。由粗模式决策过程得到的较优的多个候选帧内模式来进行率失真优化,为了缩减粗模式决策需要计算的候选模式数,在候选模式列表中根据前后帧内预测角度模式的估计损失值的差距来筛选掉部分可能性较低的候选模式实现早停止决策,从而减少需要进行率失真优化的候选模式数量,进而减少率失真优化过程的计算时间。本文提出的算法在测试序列上平均实现78.15%的编码时间缩减,BD-PSNR为-0.168 d B,BD-RATE为3.49%。展开更多
文摘By mcans of stable attractors of discret Hopfield neural network (DHNN) , anew class of nonlinear error control codes is sugsested and some relativetheorems are presented. A kind of single error control codes is also given forillustrating this new approach.
文摘Presents the fuzzy neural network optimized by hybrid coded genetic algorithm of decimal encoding and binary encoding, the searching ability and stability of genetic algorithms enhanced by using binary encoding during the crossover operation and decimal encoding during the mutation operation, and the way of accepting new individuals by probability adopted, by which a new individual is accepted and its parent is discarded when its fitness is higher than that of its parent, and a new individual is accepted by probability when its fitness is lower than that of its parent. And concludes with calculations made with an example that these improvements enhance the speed of genetic algorithms to optimize the fuzzy neural network controller.
文摘A method of assigning binary indexes to codevectors in vector quantization (VQ)system, which is called pseudo-Gray coding, is presented in this paper by constructing a kind of Hopfield neural network. Pseudo-Gray coding belongs to joint source/channel coding, which could provide a redundancy-free error protection scheme for VQ of analog signals when the binary indexes of signal codevectors are used as channel symbols on a discrete memoryless channel. Since pseudo-Gray coding is of combinatorial optimization problems which are NP-complete problems,globally optimal solutions are generally impossible. Thus, a kind of Hopfield neural network is used by constructing suitable energy function to get sub-optimal solutions. This kind of Hop field neural network is easily modified to solve simplified version of pseudo-Gray coding for single bit-error channel model. Simulating experimental results show that the method introduced here could offer good performances.
基金Supported by the Eleventh Five-Year Pre-research Project of China.
文摘A new 3D surface contouring and ranging system based on digital fringe projection and phase shifting technique is presented. Using the phase-shift technique, points cloud with high spatial resolution and limited accuracy can be generated. Stereo-pair images obtained from two cameras can be used to compute 3D world coordinates of a point using traditional active triangulation approach, yet the camera calibration is crucial. Neural network is a well-known approach to approximate a nonlinear system without an explicit physical model, in this work it is used to train the stereo vision application system to calculating 3D world coordinates such that the camera calibration can be bypassed. The training set for neural network consists of a variety of stereo-pair images and the corresponding 3D world coordinates. The picture elements correspondence problem is solved by using projected color-coded fringes with different orientations. Color imbalance is completely eliminated by the new color-coded method. Once the high accuracy correspondence of 2D images with 3D points is acquired, high precision 3D points cloud can be recognized by the well trained net. The obvious advantage of this approach is that high spatial resolution can be obtained by the phase-shifting technique and high accuracy 3D object point coordinates are achieved by the well trained net which is independent of the camera model works for any type of camera. Some experiments verified the performance of the method.
文摘针对神经网络超参数优化效果差、容易陷入次优解和优化效率低的问题,提出一种基于改进实数编码遗传算法(IRCGA)的深度神经网络超参数优化算法——IRCGA-DNN(IRCGA for Deep Neural Network)。首先,采用实数编码方式表示超参数的取值,使超参数的搜索空间更灵活;然后,引入分层比例选择算子增加解集多样性;最后,分别设计了改进的单点交叉和变异算子,以更全面地探索超参数空间,提高优化算法的效率和质量。基于两个仿真数据集,验证IRCGA-DNN的毁伤效果预测性能和收敛效率。实验结果表明,在两个数据集上,与GA-DNN(Genetic Algorithm for Deep Neural Network)相比,所提算法的收敛迭代次数分别减少了8.7%和13.6%,均方误差(MSE)相差不大;与IGA-DNN(Improved GA-DNN)相比,IRCGA-DNN的收敛迭代次数分别减少了22.2%和13.6%。实验结果表明,所提算法收敛速度和预测性能均更优,能有效处理神经网络超参数优化问题。
文摘相比于高效视频编码(high efficiency video coding,HEVC)标准,新一代编码标准多功能视频编码(versatile video coding,VVC)引入了很多新的技术,其中包括四叉树(quadtree,QT)和多类型树(multi-type tree,MTT)划分,MTT划分由HEVC中的QT划分延伸而来。新划分方法提高了压缩效率,但导致编码时间急剧增加。为了降低编码复杂度,提出了一种结合深度学习方法和MTT方向早期判决的快速帧内编码算法。首先使用轻量级的卷积神经网络(convolutional neural network,CNN)对QT和部分MTT进行预测划分,其余MTT则采用提前预测MTT划分方向的方法作进一步的优化。实验结果表明,所提方法能够大幅降低编码复杂度,相比于原始编码器的编码时间减少了74.3%,且只有3.3%的码率损失,性能优于对比的方法。
文摘为了提升高效视频编码(High Efficiency Video Coding,HEVC)帧内编码的实时性能,本文提出的方法利用了引入偶数边长与步长的卷积核以及自注意力机制的轻量级卷积网络来预测编码树单元(Coding Tree Unit,CTU)的帧内划分结构,从而减少了编码器对CTU进行四叉树递归遍历划分的编码时间。原始编码策略中粗模式决策通过基于残差经哈德曼变换的预测残差绝对值总和(Sum of Absolute Transformed Difference,SATD)的损失值来估计率失真优化过程中的率失真损失值来进行加速,但仍会耗费一定的编码时间。提出一种方法通过采样搜索的方式减少粗模式决策过程中计算的模式数,从35种模式降低到了18种模式,降低了粗模式决策过程中计算估计损失值的时间。由粗模式决策过程得到的较优的多个候选帧内模式来进行率失真优化,为了缩减粗模式决策需要计算的候选模式数,在候选模式列表中根据前后帧内预测角度模式的估计损失值的差距来筛选掉部分可能性较低的候选模式实现早停止决策,从而减少需要进行率失真优化的候选模式数量,进而减少率失真优化过程的计算时间。本文提出的算法在测试序列上平均实现78.15%的编码时间缩减,BD-PSNR为-0.168 d B,BD-RATE为3.49%。
