ID 原文 译文
58788 针对传统的Sigmoid激活函数拟合方法准确度不高、消耗大量资源等问题,提出了一种基于各层神经元值分布概率的Sigmoid函数分段线性拟合方法,以便在仅使用加法电路的情况下,提高神经网络的识别精度。首先以Sigmoid函数的二阶导数为基础,将Sigmoid函数划分为3个固定区域; In order to improve the network recognition accuracy in the low complexity condition, a piecewise linear sigmoid function approximation based on the distribution probability of the neurons’ values is proposed only with one addition circuit. The sigmoid function is first divided into three fixed regions.
58789 其次根据每层神经元值的概率变化,将每个固定区域的曲线再划分成不同数量、长度的子区域,以减少近似误差, 提高识别精度。 Second, according to the neurons’ values distribution probability, the curve in each region is segmented into sub-regions to reduce the approximation error and improve the recognition accuracy.
58790 分段线性函数斜率设为2 - n,有效地降低了Sigmoid函数的硬件实现复杂度。 The slope of the piecewise linear function is set as 2-n, effectively reducing the hardware implementation complexity.
58791 最后,设计了拟合函数的硬件电路结构,并在Xilinx FPGA-XC7A200T上实现MNIST手写数字识别,对所提出的拟合方法进行验证。 Experiments performed on Xilinx’s FPGA-XC7A200T implement the MNIST handwritten digits recognition.
58792 实验结果表明,该方法的识别准确率在深度神经网络中约达到97.45%,而在卷积神经网络中约达到98.42%。与其他仅使用加法电路的拟合方法对比,准确率分别提高了约0.84%和0. 57%。 The results show that the proposed method achieves a 97. 45% recognition accuracy in a deep neural network and 98. 42% in a convolutional neural network, up to 0. 84% and 0. 57% higher than other approximation methods only with one addition circuit.
58793 为了提高识别含噪图像的能力,提出一种基于深度残差学习的含噪图像轮廓重建方法。 In order to improve the recognition ability of noisy images, a method of contour reconstruction based on depth residuals learning is proposed.
58794 采用锐化模板匹配技术进行含噪图像信息的增强处理;利用图像的局部灰度信息构建图像的边缘活动轮廓模型;采用活动轮廓套索方法进行图像高分辨重构; The sharpening template matching technique is used to enhance the noisy image information, the local gray level information on the image is used to construct the edge active contour model of the image, and the active contour lasso method is used to reconstruct the image with a high resolution.
58795 提取含噪图像的局部灰度能量项与局部梯度能量项特征量,构建卷积神经网络分类器进行特征分类; The feature quantities of local gray energy and local gradient energy of the noisy image are extracted, and a convolutional neural network classifier is constructed to classify the features. The learning depth of the learning convolutional neural network is judged by combining the similarity of the gray histogram of the image.
58796 结合图像灰度直方图的相似性判断学习的卷积神经网络的学习深度,提升图像细节信息的分辨能力,实现含噪图像的轮廓高分辨重建。 The resolution ability of image detail information is improved, and the contour high resolution reconstruction of the noisy image is realized.
58797 仿真结果表明,采用该方法进行含噪模糊图像重建的分辨能力较高,输出峰值信噪比较高,有效地提升了图像的识别能力。 Simulation results show that the proposed method has a high resolution and a high peak signal to noise ratio (PSNR), which improves the recognition ability of the image effectively.