Presentation attack detection based on two-stream vision transformers with self-attention fusion

Aiming at the performance degradation of the existing presentation attack detection methods due to the illumination variation, a two-stream vision transformers framework (TSViT) based on transfer learning in two complementary spaces is proposed in this paper. The face images of RGB color space and multi-scale retinex with color restoration (MSRCR) space are fed to TSViT to learn the distinguishing features of presentation attack detection. To effectively fuse features from two sources (RGB color space images and MSRCR images), a feature fusion method based on self-attention is built, which can effectively capture the complementarity of two features. Experiments and analysis on Oulu-NPU, CASIA-MFSD, and Replay-Attack databases show that it outperforms most existing methods in intra-database testing and achieves good generalization performance in cross-database testing.     

Spectral radiance reconstruction from trichromatic camera responses based on orthogonal test and regularized algorithm

Reconstruction of spectral information based on multi‐channel image system is a significant problem in color reproduction, detection, and recognition. A spectral radiance reconstruction from trichromatic digital camera responses is researched in this article. The mapping relationship between the trichromatic imaging system response and the incident spectral radiance is analyzed. Then, in order to remove the ill‐posedness of the problem, a regularized constraint solution model of spectral radiance reconstruction matrix is established. And the spectral radiance can be reconstructed by spectral radiance reconstruction matrices and trichromatic imaging system response. Finally, the spectral radiance reconstruction matrix is estimated by the system radiometric calibration experiment. The input radiance is offered by a LCD display. A 3‐factor and 9‐level orthogonal test is designed for the calibration experiment, and a test set of 24 colors is used for precision analysis. The results show that the average relative mean error of our method is 8.69%, it is lower than that of Wiener filtering method by 2.84%. The method can reconstruct spectral radiance information effectively.     

A customized low-rank prior model for structured cartoon–texture image decomposition

The mathematical characterization of the texture component plays an instrumental role in image decomposition. In this paper, we are concerned with a low-rank texture prior based cartoon–texture image decomposition model, which utilizes a total variation norm and a global nuclear norm to characterize the cartoon and texture components, respectively. It is promising that our decomposition model is not only extremely simple, but also works perfectly for globally well-patterned images in the sense that the model can recover cleaner texture (or details) than the other novel models. Moreover, such a model can be easily reformulated as a separable convex optimization problem, thereby enjoying a splitting nature so that we can employ a partially parallel splitting method (PPSM) to solve it efficiently. A series of numerical experiments on image restoration demonstrate that PPSM can recover slightly higher quality images than some existing algorithms in terms of taking less iterations or computing time in many cases.     

A new linearly constrained minimum variance beamformer for reconstructing EEG sparse sources

Brain source imaging based on EEG aims to reconstruct the neural activities producing the scalp potentials. This includes solving the forward and inverse problems. The aim of the inverse problem is to estimate the activity of the brain sources based on the measured data and leadfield matrix computed in the forward step. Spatial filtering, also known as beamforming, is an inverse method that reconstructs the time course of the source at a particular location by weighting and linearly combining the sensor data. In this paper, we considered a temporal assumption related to the time course of the source, namely sparsity, in the Linearly Constrained Minimum Variance (LCMV) beamformer. This assumption sounds reasonable since not all brain sources are active all the time such as epileptic spikes and also some experimental protocols such as electrical stimulations of a peripheral nerve can be sparse in time. Developing the sparse beamformer is done by incorporating L1-norm regularization of the beamformer output in the relevant cost function while obtaining the filter weights. We called this new beamformer SParse LCMV (SP-LCMV). We compared the performance of the SP-LCMV with that of LCMV for both superficial and deep sources with different amplitudes using synthetic EEG signals. Also, we compared them in localization and reconstruction of sources underlying electric median nerve stimulation. Results show that the proposed sparse beamformer can enhance reconstruction of sparse sources especially in the case of sources with high amplitude spikes.     

基于VMD和改进TCN的短期光伏发电功率预测

光伏发电功率存在波动性,且光伏出力易受各种气象特征影响,传统TCN网络容易过度强化空间特性而弱化个体特性。针对上述问题,文中提出一种基于VMD和改进TCN的短期光伏发电功率预测模型。通过VMD将原始光伏发电功率时间序列分解为若干不同频率的模态分量,将各个模态分量以及相对应的气象数据输入至改进TCN网络进行建模学习。利用中心频率法确定VMD的最优分解模态分解个数。在传统TCN预测模型的基础上,使用DropBlock正则化取代Dropout正则化以达到抑制卷积层中信息协同的效果,并引入注意力机制自主挖掘并突出关键气象输入特征的影响,量化各气象因素对光伏发电的影响,从而提高预测精度。以江苏省某光伏电站真实数据为例进行仿真实验,结果表明所提预测方法的RMSE为0.62 MW,MAPE为2.03%。     

酸性尾矿废弃地生态修复技术进展

酸性尾矿废弃地废石含有硫化物，在雨水的淋溶作用下会产出硫酸，加剧废弃地的酸化，增加企业环保成本，而生态修复工作就是从源头上控制废弃地的酸化。根据尾矿废弃地酸化机理，结合国内外最新研究进展，分析了生态修复技术现状；概述了隔离覆盖、改良剂改良、植物修复、微生物法等多种处理方法在实践中的适应性及优缺点，并分别介绍了隔离覆盖-植物修复技术、原位改良-植物修复技术在Reden煤矿、永平铜矿的成功应用案例。针对酸性尾矿废弃地的特点，单一处理方法具有较大的局限性，需要采用联合工艺进行治理，隔离覆盖-植物修复技术、原位改良-植物修复技术因见效快、效果好、操作简单，势必成为将来研究的主要方向。     

Application of fuzzy image restoration in criminal investigation

The advancement of science and technology has a positive effect on the development of law disciplines. The development of algorithms and artificial intelligence also has a certain impact on judicial practice. Image restoration is a significant technique in image processing. It aims to objectively restore the content or quality of the original image from the degraded image. Image degradation is always generated in image transmission, such as distortion, blur. In modern video surveillance system, image restoration is significant for criminal investigation. However, image restoration based on conventional filter algorithms cannot achieve satisfactory performance. Thus, we first introduce the image restoration algorithms based on different degradation model. Then, we propose some applications of fuzzy image restoration in criminal investigation. We conduct experiments on both degraded images and videos and experimental results have shown the effectiveness of fuzzy image restoration applying to the criminal investigation.     

船体结构冰载荷反演方法及试验验证

冰载荷是影响船舶冰区航行期间结构安全的重要环境载荷。船舶的冰压监测通常采用应变传感器，合理地布放传感器是识别冰载荷的基础。通过对比船体外板结构试验中的冲击载荷和不同测试方案下的应变信号，确定了最佳应变传感器布放方案；采用Green核函数方法建立了船体外板结构应变冲击载荷间的响应关系，并对采集信号在噪声影响下反演的不适定性进行了分析；采用Tikhonov正则化方法克服了载荷反演过程中出现的数值不稳定问题；最后将试验中的响应用到载荷识别分析中，反演的载荷可以较为准确地反映冲击载荷的时域特征并且载荷识别精度良好。     

自适应融合层级特征的混合退化图像复原算法

多种退化类型混合的图像比单一类型的退化图像降质更严重,很难建立精确模型对其复原,研究端到端的神经网络算法是复原的关键.现有的基于操作选择注意力网络的算法(operation-wiseattentionnetwork,OWAN)虽然有一定的性能提升,但是其网络过于复杂,运行较慢,复原图像缺乏高频细节,整体效果也有提升的空间.针对这些问题,提出一种基于层级特征融合的自适应复原算法.该算法直接融合不同感受野分支的特征,增强复原图像的结构;用注意力机制对不同层级的特征进行动态融合,增加模型的自适应性,降低了模型冗余;另外,结合L1损失和感知损失,增强了复原图像的视觉感知效果.在DIV2K,BSD500等数据集上的实验结果表明,该算法无论是在峰值信噪比和结构相似性上的定量分析,还是在主观视觉质量方面,均优于OWAN算法,充分证明了该算法的有效性.