To solve the problem of insufficient facial expression feature extraction by the VGG16 network, resulting in low recognition accuracy, an improved VGG16 network is proposed. Firstly, take the convolutional layer with the same channel in the VGG16 network as a block and divide it into five blocks.Then use the feature fusion method, the features extracted from the last three blocks of VGG16 network are fused to enhance the extraction of facial expression features. Finally, the attention mechanism SGE (spatial group enhance) module is introduced to promote the network to extract features that are conducive to facial expression recognition. At the same time, the three fully connected layers of the VGG16 network are changed to one fully connected layer to directly output the classification results, which can reduce the number of parameters while ensuring the accuracy of recognition. Experimental results show that the improved VGG16 network achieves 86.701% accuracy on RAF-DB and 56.881% accuracy on SFEW, which is higher than the original network.

为解决现有喷码机无法进行远程操控及喷码机不够智能化的问题,设计了一种双ARM嵌入式喷码机系统,该系统以ARM9系列Exynos4412为主处理器,搭配ARM Cortex-M3内核的STM32为辅处理器,完成了电源供电电路、墨滴充电电路、相位检测电路以及高压偏转电路的硬件设计;根据系统功能所需,裁剪、定制出一套适配该系统的Linux嵌入式操作系统;同时在所设计的操作系统的基础上,设计出基于Web服务器boa网络通信并可在远程PC端进行操控的喷码机软件管理系统;经实验测试,结果表明系统的硬件设计达到了系统要求,远程操作控制系统反应灵敏,使得喷码机更加智能化.

For the insufficient extraction of facial expression features by convolutional neural network VGG16, an improved VGG16-NFB network model is proposed to extract facial expression features more fully, so as to improve the accuracy of facial expression recognition. Firstly, VGG16 is used to optimize the network structure, and the part before the classifier is used as the feature extractor to extract the features of facial expressions initially, and the redundant two full connection layers in the classifier are deleted. Then, a feature enhancement module FEM is designed, in which the representative batch normalization method can calibrate the feature information effectively, make the feature distribution more stable, and further enhance the network's ability to extract features from facial expressions. Finally, the global average pooling is used to extract the features of each level in the feature enhancement module, and the bridge attention module BAM is introduced to fuse the features of each level, which is used to generate attention weights to further enhance the facial expression features with high discrimination. The experimental results show that the improved VGG16-NFB network achieves 86.8% and 87.5% recognition accuracy on the RAF-DB and FERPlus facial expression datasets, respectively, which is 2.22% and 1.61% higher than the original network.

Facial Expression Recognition(FER), an important method in the field of human-computer interaction, is widely used for intelligent medical treatment and in public security lie detection systems, vehicle safety systems, and other applications.However, existing FER methods are problematic in that they are plagued by parameter redundancy, high computational cost, feature expression bottlenecks, and so on.This study proposes an FER method based on an improved lightweight Rank Expansion Network(ReXNet).By constructing improved ReXNet to exact facial expression features, solves the feature expression bottleneck by decreasing the number of parameters and enhancing attention to local features of expression to produce high-level features of expression.At the same time, a Coordinate Attention(CA) module is integrated, and the location information is embedded into the channel attention.These improvements enable the model to accurately locate and identify the features of interest, and to establish a long-term relationship between location information and local features, thereby reducing the computational overhead.On this basis, a refinement module is introduced to the improved model.The module refines the classification results by using the category context information to enhance the differentiation between classes to ultimately improve the accuracy of FER.The experimental results show that the accuracy of FER on the RAF-DB and FERPlus datasets reaches 88.43% and 88.8%, respectively.Compared with VGG16-PLD, SHCNN, ResNet+VGG, and other methods, the proposed method has higher accuracy and superior robustness.

As it is difficult to extract effective features in facial expression recognition and the high similarity between categories and easy confusion lead to low accuracy of facial expression recognition, a facial expression recognition method based on anti-aliasing residual attention network is proposed. First, in view of the problem that the traditional subsampling method can easily cause the loss of expression discriminative features, an anti-aliasing residual network is constructed to improve the feature extraction ability of expression images and enhance the representation of expression features, enabling more effective global facial expression information to be extracted.At the same time, the improved channel attention mechanism and label smoothing regularization strategy are used to enhance the attention to the local key expression regions of the face: the improved channel attention focuses on the highly discriminative expression features and suppresses the weight of non-expressive regions, so as to locate more detailed local expression regions in the global information extracted by the network, and the label smoothing technology corrects the prediction probability by increasing the amount of information of the decision-making expression category, avoiding too absolute prediction results, which reduces misjudgment between similar expressions. Experimental results show that, the recognition accuracies of this method on the facial expression datasets RAF-DB and FERPlus reach 88.14% and 89.31%, respectively.Compared with advanced methods such as DACT and VTFF, this method has better performance. Compared with the original residual network, the accuracy and robustness of facial expression recognition are effectively improved.

In the process of facial expression recognition by convolutional neural network, aiming at the problem that the complex background interferes with the extraction of expression features, a simple face cropping strategy is proposed. First, the critical facial expression regions are calculated by face alignment and landmarks detection, thus the background influence outside the facial expression region is reduced, and then the convolutional neural network is used to further extract expression features and enable expression classification. The experimental results show that the facial expression recognition effect is significantly improved by the proposed method, and the recognition accuracy on the facial expression datasets JAFFE and CK+ reaches 90.48% and 96.67%, respectively.

A face expression recognition method based on an improved rank expansion network is proposed to address the problem that when traditional convolutional neural networks recognize facial expressions, the network has difficulty in fully extracting the important expression features leading to low recognition accuracy. First, based on the idea of migration learning, the pre-training parameters of large-scale image sets are used to initialize the rank expansion network and accelerate the training of the model. Then, the face expression features are further enriched by fusing the underlying detailed features and the high-level semantic features at different scales. Finally, a stepped attention module is embedded before feature fusion to construct the contextual relationships of semantic features from three different dimensions to strengthen the expression of important features. The experimental results show that the improved network achieves 88.07% and 88.68% accuracy on the face expression datasets RAF-DB and FERPlus, which is a significant improvement over the original network.

A facial expression recognition method with improved residual network is proposed to address the problems of low recognition accuracy and loss of detailed features in the discriminative task due to inappropriate pooling in current facial expression recognition models. First, multi-scale importance pooling with adaptively learned input feature weights is proposed to improve the model's ability to retain and recognize global features. Second, asymmetric convolution is used to augment the convolution kernels to improve the characterization of local features, and the set of asymmetric convolution kernels trained to convergence is fused and equivalently converted to the original network structure to avoid introducing additional parameter optimization in the inference stage. Finally, the expression recognition accuracies of 85.53% and 72.72% were achieved in two publicly available datasets, RAF-DB and FER2013, respectively, and the experimental results demonstrate the effectiveness of the proposed method with certain application prospects.

针对轻量级网络在复杂环境下对面部表情的特征提取不够充分、泛化能力不足以及单标签数据集无法有效描述复杂情感倾向所带来的歧义表情等问题，提出了一种结合改进ShuffleNet与标签分布学习的人脸表情识别方法。在不大量增加计算复杂度的前提下，为了避免模型的过拟合，设计了新的输出模块对ShuffleNet模型进行改进；为了增强模型对人脸表情图像重要局部细节特征的提取能力，设计了并行的深度卷积残差模块，实现了局部与全局特征的融合。为了减少由歧义表情对识别性能所带来的不利影响，通过标签分布学习方法，在不引入额外信息的前提下，充分利用数据集原本信息生成标签分布，并重新训练改进后的ShuffleNet模型。实验结果表明，在RAF-DB、AffectNet-7和AffectNet-8数据集上分别达到了87.15%、62.05%和58.49%的准确率，同时参数量和计算量均保持在较低水平，利于其在实际生产中应用。

At present, facial expression recognition is widely used and is an important part of human-computer interaction. With the rapid development of deep learning, facial expression recognition technology will be more widely used. However, when facial expression recognition technology is applied to large-scale facial expression database, the accuracy of facial expression recognition is low and it is easy to produce defects such as over fitting. In order to optimize the deep learning network and further improve the accuracy of facial expression recognition, this paper proposes an improved network model combined with resnet18. The model replaces the relu activation function after the accumulation layer of the backbone network with the mish activation function. After the feature extraction of facial expression pictures, softpool is used to improve the information loss caused by down sampling, which is conducive to distinguishing similar key points, So as to improve the accuracy of facial expression recognition. The improved network model is verified on raf-db, CK + and Jaffe data sets, and the human face expression recognition rates reach 87.54

目前大规模人脸表情识别的主要问题在于不确定性,这些不确定性来源于模棱两可的面部表情、低质量的脸部图像和标注者的主观性.为此,文中提出一种基于RepVGG-A0改进后的网络模型.该模型引入有效通道注意力机制,即在卷积层和ReLU激活函数之间插入ECA通道注意力模块,在特征提取后引入加权模块来预计样本的权重,对于不确定的样本给予的权重较小,并采用重新标签的方法对低权重的人脸表情图片重新给予伪标签,目的是使修改过的样本在下一次训练中获得高权重,从而提高人脸表情识别率.最后,在RAF-DB和FER-2013数据集上进行实验验证.结果表明,文中改进模型的人脸表情识别率分别达到88.90％和75.61％,说明该方法对人脸表情识别具有有效性.

人的表情包含大量信息,可用于显示人的很多情感状态,例如疲劳和疼痛的表情等.卷积神经网络(CNN)是一种识别人脸表情的有效方法,它可以同时执行特征提取和分类,并可以自动发现数据中的多个表情特点.针对卷积神经网络参数大以及传统表情识别方法准确率不高的问题,提出一种基于改进的稠密卷积神经网络的面部表情识别模型.首先通过使用Gabor滤波器初始化第一层卷积层;然后采用一种对数线性函数(LLU)进行网络优化,该模型中的特征重用和参数压缩技术提高了网络的学习能力,大大减少了模型参数;最后基于此模型设计了一个表情识别系统,该系统能够准确地识别照片上的表情和在线识别人脸表情.实验结果表明,该模型可以显著提高三个表情数据集的准确率,并能很好地识别人脸表情.

Facial expressions are the main external manifestations of human emotions. Facial expression recognition technology can be used in medical, investigation, education and other application scenarios. Aiming at the disadvantages of slow convergence speed and low recognition accuracy of traditional neural networks, in order to optimize the network more efficiently and improve the recognition rate of facial expressions, a facial expression recognition method based on improved convolutional neural networks is proposed.First, the deep residual network ResNet18 is improved, and then the improved ResNet18 network is used to extract global expression features of face images, and then the self-attention weighting module is introduced to calculate the expression features of each face image and output one The corresponding weights are used to weight the loss function. Finally, the two public expression data sets of CK+ and RAF-DB are used to experimentally verify the method in the article. The accuracy of facial expression recognition reaches 98.89% and respectively. 87.13%, compared with the original deep residual network ResNet18 model and other types of network models, the facial expression recognition rate has been significantly improved, which proves that this method can effectively improve the accuracy of facial expression recognition and has certain application value.

针对已有卷积神经网络在手势识别过程中精度不高的问题,提出了一种双通道卷积神经网络的特征融合与动态衰减学习率相结合的复合型手势识别方法.通过两个相互独立的通道进行手势图像的特征提取,首先使用SENet(Squeeze-and-Excitation Networks)构成的第一通道提取全局特征,然后使用RBNet(Residual Block Networks)构成的第二通道提取局部特征,并将全局特征和局部特征进行通道维度上的融合.同时,利用动态衰减的学习率训练双通道网络模型.与其他卷积神经网络模型的对比实验结果表明,提出的复合型手势识别方法的手势识别率高,参数数量少,适用于不同手势图像数据集的识别.

由于受到光照和成像设备等条件因素的影响,采集到的单帧人脸图像分辨率低,无法进行准确人脸识别,所以需要图像超分辨率重建.而利用SRGAN模型在进行人脸超分辨率重建过程中,易出现梯度消失或爆炸的问题,严重影响了重建图像的精度和质量.针对上述问题,提出了基于生成对抗网络的改进人脸超分辨率重建算法,在SRGAN结合WGA-N的基础上引入Wasserstein散度,并将其最大化得到最优化标量函数T,实现去掉Lipschit-z约束能够直接得到Wassertein距离,并通过最小化Wassertein距离得到生成网络的目标函数,最终改进的模型提高了重建图像的质量.实验结果表明,该方法能够生成高分辨率的人脸图像,在主观视觉和客观评价指标均同比优于DRCN、FSRCNN、SRGAN_WGAN、VDSR和DRRN模型.

When the SSD network uses an independent feature layer to detect objects, there is no connection between the layers, which leads to the problem of insufficient expression of contextual features. This paper proposes a pedestrian detection algorithm based on improved SSD network. The algorithm uses cross-layer feature adaptive fusion, and combines residual channel attention modules with different convolution rates of holes. While increasing the receptive field, this algorithm enhances important features and weakens unimportant features. Make the extracted features more directional, thereby improving the accuracy of pedestrian detection. Experiment on the improved network and algorithm on the INRIA pedestrian detection dataset, and the mixed pedestrian dataset extracted from the COCO dataset and the Crowd human dataset. The experimental results show that the average precision of pedestrian detection in the two datasets is improved by 1.7% and 4.0% respectively compared with the original network.

A DNAzyme-mediated target-initiated rolling circle signal amplification strategy based on a microchip platform was developed for detecting apurinic/apyrimidine endonuclease 1 (APE1) at the single-cell level. This strategy was applied to assays of lysate samples from HL-7702, HeLa and MCF-7 cells, with a detection limit of lower than 1 HeLa cell.

针对三轴增稳云台伺服系统非线性特性,以及PD控制抗扰能力差,自抗扰控制器由于参数众多而导致整定过程耗时且费力的缺陷,本文利用BP神经网络的全局逼近能力和自我学习能力,将其与自抗扰控制器组成复合控制器,对自抗扰控制器的所有关键参数进行自整定寻优,应用于含Stribeck摩擦模型的三轴增稳云台伺服系统.仿真结果表明:该方法用于自动整定参数可行有效,与PD控制和参数固定的常规自抗扰控制器相比,具有更高的控制精度和更强的抗扰能力,对提高增稳云台的性能具有较好的应用价值.

在激光打标系统中,逐点比较插补法是一种普遍使用的算法.这种算法在控制激光运动轨迹和对图像轨迹的拟合过程中,起到极其重要的作用.由于逐点比较插补法具有运算直观,脉冲输出均匀,调节方便等特点,所以是一种较为适合打标机系统应用的插补算法.但是传统的逐点比较插补算法存在精度低和速度慢等问题.为了改善该算法的性能,对传统逐点比较插补算法进行了改进,并进行了计算和模拟仿真.结果 表明改进比较插补算法在精度和速度上都有明显提高.

利用手持云台对黑白照/图像进行拍摄再对其进行上色是耗时耗力的工作,为了提高对黑白照片/图像上色的效率和视觉效果,利用深度学习的卷积神经网络提出了一种基于改进生成对抗网络的上色算法.采用了原模型pix2pix(image-to-image translation with conditional generative adversarial networks)的U型结构的生成器并在其中引入了自注意力机制提高输出图像的色彩多样性;其次是使用实例归一化对网络结构作进一步的优化处理,生成器和判别器在网络训练期间不断进行相互对抗学习,同时模型不断学习并优化黑白照片/图像到对应彩色图像的映射关系;最后实现了对黑白照片/图像的自动化上色,同时使用客观量化指标(MSE、PSNR和SSIM)对实验结果进行评价.实验结果表明:该算法能快速有效实现对黑白照片/图像的无监督上色,同时保持良好的视觉效果.