基于正交回归和特征加权的脑电情感特征选择方法

颅内容积传导效应导致大量脑电特征之间具有高度相关性,而这些高度相关的脑电特征无法为情感识别提供额外的有用信息,并且会降低基于脑电信号的情感识别效率.为了去除冗余信息和挑选有判别力的脑电特征,本文提出了一种基于正交回归和特征加权的脑电情感特征选择方法.与传统特征选择方法相比,该方法利用正交回归在脑电特征映射空间中保留更多的判别信息,更加适合于非线性和非平稳脑电信号的分析处理.为了验证所提出方法的性能,我们采集了由视频诱发的多通道脑电情感数据,并将所提出方法与4种常用的脑电特征选择方法进行了比较.实验结果证明了本文所提出方法能有效降低脑电特征集内冗余信息,并挑选出具有判别力的脑电特征子集.此外,通过分析由该方法所挑选的脑电特征类型,我们发现中心频率特征是最具判别力的脑电情感特征.该发现将为未来脑电情感特征提取研究提供新的思路.

基于神经网络的生成式三维数字人研究综述:表示、渲染与学习

随着人工智能技术的高速发展,计算机视觉与图形学等相关学科的交叉融合掀起了一场数字人生成技术的新革命,人类进入“元宇宙”等数字空间的梦想正逐渐变为现实.面对大规模三维数字人的生产需求,传统图形学建模方法建模过程繁琐,周期冗长,阻碍了虚拟数字人的普及和应用,而利用生成式人工智能技术产生高拟真、规模化的虚拟数字人正逐渐成为研究热点.为了深入了解三维数字人技术的研究现状与挑战,本文从生成式模型的视角对数字人技术进行了系统性梳理,并总结了其中的3个关键步骤:表示、渲染与学习.随后,对显式及隐式的表示方法进行总结,对传统渲染与神经网络渲染的成像方式进行归纳,并概括了相应的模型学习方法.最后,本文对三维数字人的典型应用进行分析,并对当前挑战与未来发展方向进行总结和展望.

Auxiliary Diagnosis of Papillary Thyroid Carcinoma Based on Spectral Phenotype

Thyroid cancer,a common endocrine malignancy,is one of the leading death causes among endocrine tumors.The diagnosis of pathological section analysis suffers from diagnostic delay and cumbersome operating procedures.Therefore,we intend to construct the models based on spectral data that can be potentially used for rapid intraoperative papillary thyroid carcinoma(PTC)diagnosis and characterize PTC characteristics.To alleviate any concerns pathologists may have about using the model,we conducted an analysis of the used bands that can be interpreted pathologically.A spectra acquisition system was first built to acquire spectra of pathological section images from 91 patients.The obtained spectral dataset contains 217 spectra of normal thyroid tissue and 217 spectra of PTC tissue.Clinical data of the corresponding patients were collected for subsequent model interpretability analysis.The experiment has been approved by the Ethics Review Committee of the Wuhu Hospital of East China Normal University.The spectral preprocessing method was used to process the spectra,and the preprocessed signal respectively optimized by the first and secondary informative wavelengths selection was used to develop the PTC detection models.The PTC detection model using mean centering(MC)and multiple scattering correction(MSC)has optimal performance,and the reasons for the good performance were analyzed in combination with the spectral acquisition process and composition of the test slide.For model interpretable analysis,the near-ultraviolet band selected for modeling corresponds to the location of amino acid absorption peak,and this is consistent with the clinical phenomenon of significantly lower amino acid concentrations in PTC patients.Moreover,the absorption peak of hemoglobin selected for modeling is consistent with the low hemoglobin index in PTC patients.In addition,the correlation analysis was performed between the selected wavelengths and the clinical data,and the results show:the reflection intensity of selected wavelengths in normal cells has a moderate correlation with cell arrangement structure,nucleus size and free thyroxine(FT4),and has a strong correlation with triiodothyronine(T3);the reflection intensity of selected bands in PTC cells has a moderate correlation with free triiodothyronine(FT3).

Palmprint Phenotype Feature Extraction and Classification Based on Deep Learning

Palmprints are of long practical and cultural interest.Palmprint principal lines,also called primary palmar lines,are one of the most dominant palmprint features and do not change over the lifespan.The existing methods utilize filters and edge detection operators to get the principal lines from the palm region of interest(ROI),but can not distinguish the principal lines from fine wrinkles.This paper proposes a novel deep-learning architecture to extract palmprint principal lines,which could greatly reduce the influence of fine wrinkles,and classify palmprint phenotypes further from 2D palmprint images.This architecture includes three modules,ROI extraction module(REM)using pre-trained hand key point location model,principal line extraction module(PLEM)using deep edge detection model,and phenotype classifier(PC)based on ResNet34 network.Compared with the current ROI extraction method,our extraction is competitive with a success rate of 95.2%.For principal line extraction,the similarity score between our extracted lines and ground truth palmprint lines achieves 0.813.And the proposed architecture achieves a phenotype classification accuracy of 95.7%based on our self-built palmprint dataset CAS_Palm.

Identification of various food residuals on denim based on hyperspectral imaging system and combination optimal strategy

As the science and technology develop,crime methods and scenes have become increasingly complex and diverse.Trace evidence analysis has become amore and more important criminal investigation technology and liquid is the main form of trace evidence.Food can provide not only energy,but clues to solve crimes.In this study,we build a hyperspectral imaging system to detect liquid residue traces,including apple juice,coffee,cola,milk and tea,on denims with light,middle and dark colors.The obtained hyperspectral images are first subjected to spectral calibration and hyperspectral data pretreatment.Subsequently,Partial Least Squares(PLS)is applied to select the informative wavelengths from the preprocessed spectra.For modeling phase,the combination optimal strategy,support vector machine(SVM)combined with random forest(RF),is developed to establish classification models.The experimental results demonstrate that the combination optimal model can achieve TPR,FPR,Precision,Recall,F1,and AUC of 83.5%,2.30%,79.7%,83.5%,81.6%,and 94.7%for classifying fabrics contaminated by various food residuals.With respect to the classification of liquid and fabric types,the combination optimalmodel also yields satisfactory classification performance.In future work,wewill expand the types of liquid,and make appropriate adjustment to algorithms for improving the robustness of classification models.This research may play a positive role in the construction of a harmonious society.

Psycho-visual modulation based information display:introduction and survey

Industry and academia have been making great efforts in improving refresh rates and resolutions of display devices to meet the ever increasing needs of consumers for better visual quality.As a result,many modem displays have spatial and temporal resolutions far beyond the discern capability of human visual systems.Thus,leading to the possibility of using those display-eye redundancies for innovative usages.Tempo-ral/spatial psycho-visual modulation(TPVM/SPVM)was proposed to exploit those redundancies to generate multiple visual percepts for different viewers or to transmit non-visual data to computing devices without affecting normal viewing.This paper reviews the STPVM technology from both conceptual and algorithmic perspectives,with exemplary applications in multiview display,display with visible light communication,etc.Some possible future research directions are also identified.