Attention Guided Food Recognition via Multi-Stage Local Feature Fusion

The task of food image recognition,a nuanced subset of fine-grained image recognition,grapples with substantial intra-class variation and minimal inter-class differences.These challenges are compounded by the irregular and multi-scale nature of food images.Addressing these complexities,our study introduces an advanced model that leverages multiple attention mechanisms and multi-stage local fusion,grounded in the ConvNeXt architecture.Our model employs hybrid attention(HA)mechanisms to pinpoint critical discriminative regions within images,substantially mitigating the influence of background noise.Furthermore,it introduces a multi-stage local fusion(MSLF)module,fostering long-distance dependencies between feature maps at varying stages.This approach facilitates the assimilation of complementary features across scales,significantly bolstering the model’s capacity for feature extraction.Furthermore,we constructed a dataset named Roushi60,which consists of 60 different categories of common meat dishes.Empirical evaluation of the ETH Food-101,ChineseFoodNet,and Roushi60 datasets reveals that our model achieves recognition accuracies of 91.12%,82.86%,and 92.50%,respectively.These figures not only mark an improvement of 1.04%,3.42%,and 1.36%over the foundational ConvNeXt network but also surpass the performance of most contemporary food image recognition methods.Such advancements underscore the efficacy of our proposed model in navigating the intricate landscape of food image recognition,setting a new benchmark for the field.

A Local Contrast Fusion Based 3D Otsu Algorithm for Multilevel Image Segmentation

To overcome the shortcomings of 1 D and 2 D Otsu’s thresholding techniques, the 3 D Otsu method has been developed.Among all Otsu’s methods, 3 D Otsu technique provides the best threshold values for the multi-level thresholding processes. In this paper, to improve the quality of segmented images, a simple and effective multilevel thresholding method is introduced. The proposed approach focuses on preserving edge detail by computing the 3 D Otsu along the fusion phenomena. The advantages of the presented scheme include higher quality outcomes, better preservation of tiny details and boundaries and reduced execution time with rising threshold levels. The fusion approach depends upon the differences between pixel intensity values within a small local space of an image;it aims to improve localized information after the thresholding process. The fusion of images based on local contrast can improve image segmentation performance by minimizing the loss of local contrast, loss of details and gray-level distributions. Results show that the proposed method yields more promising segmentation results when compared to conventional1 D Otsu, 2 D Otsu and 3 D Otsu methods, as evident from the objective and subjective evaluations.

Multi-Sensor Data Fusion Technologies for Blanket Jamming Localization

The localization of the blanket jamming is studied and a new method of solving the localization ambiguity is proposed. Radars only can acquire angle information without range information when encountering the blanket jamming. Netted radars could get position information of the blanket jamming by make use of radars' relative position and the angle information, when there is one blanket jamming. In the presence of error, the localization method and the accuracy analysis of one blanket jamming are given. However, if there are more than one blanket jamming, and the two blanket jamming and two radars are coplanar, the localization of jamming could be error due to localization ambiguity. To solve this confusion, the Kalman filter model is established for all intersections, and through the initiation and association algorithm of multi-target, the false intersection can be eliminated. Simulations show that the presented method is valid.

An Indoor Pedestrian Localization Algorithm Based on Multi-Sensor Information Fusion

For existing indoor localization algorithm has low accuracy, high cost in deployment and maintenance, lack of robustness, and low sensor utilization, this paper proposes a particle filter algorithm based on multi-sensor fusion. The pedestrian’s localization in indoor environment is described as dynamic system state estimation problem. The algorithm combines the smart mobile terminal with indoor localization, and filters the result of localization with the particle filter. In this paper, a dynamic interval particle filter algorithm based on pedestrian dead reckoning (PDR) information and RSSI localization information have been used to improve the filtering precision and the stability. Moreover, the localization results will be uploaded to the server in time, and the location fingerprint database will be built incrementally, which can adapt the dynamic changes of the indoor environment. Experimental results show that the algorithm based on multi-sensor improves the localization accuracy and robustness compared with the location algorithm based on Wi-Fi.

Cloned s-Lap Gene Coding Area, Expression and Localization of s-Lap/GFP Fusion Protein in Mammal Cells

s-Lap is a new gene sequence from pig retinal pigment epithelial(RPE) cells, which was found and cloned in the early period of apoptosis of RPE cells damaged with visible light. We cloned the coding area sequence of the novel gene of s-Lap and constructed its recombinant eukaryotic plasmid pcDNA3.1-GFP/s-lap with the recombinant DNA technique. The expression and localization of s-lap/GFP fusion protein in CHO and B_~16 cell lines were studied with the instantaneously transfected pcDNA3.1-GFP/s-lap recombinant plasmid. ~s-Lap/GFP fusion protein can be expressed in CHO and B_~16 cells with a high rate expression in the nuclei.

基于前后景分割的图像情感分析

图像是生活中重要的信息源之一,对其所表达的内容进行细节分析,可以更充分地利用信息资源。随着信息化的快速发展,针对图像模态开展情感分析工作已成为目前研究的一大热点。图像情感分析的主要环节依次为:情感特征提取、情感空间的选择、特征融合和情感识别分类。现有的大部分图像情感分析工作以图像整体为单位进行输入,未能充分发挥图像中局部特征的情感作用。如果不能对图像的全局特征和局部特征作出区分,当图像出现清晰度不高、背景噪声较多等问题时,图像的全局特征就会变得较为敏感,特征提取和识别工作将会受到严重干扰,对情感分析的准确性产生一定影响。针对目前图像情感分析存在的不足,提出一种基于前后景分割的图像情感分析方法。该方法以YOLOv5为框架,引入ConvNeXt模块和AFF模块,分别进行特征提取和注意力融合。实验结果表明,与目前比较流行的几种图像情感分析方法相比,该方法对于包含更多情感信息和语义信息的场景更为适用,性能也有所提升。

融合注意力和上下文信息的遥感图像小目标检测算法

对多尺度的遥感图像进行小目标检测时,基于深度学习的目标检测算法容易出现误检和漏检的情况。这是因为此类算法的特征提取模块进行了多次的下采样操作;而且未能根据不同类别、不同尺度的目标关注所需的上下文信息。为了解决该问题,提出一种融合注意力和上下文信息的遥感图像小目标检测算法ACM-YOLO(Attention-Context-Multiscale YOLO)。首先,应用细粒度的查询感知稀疏注意力以减少小目标特征信息的丢失,从而避免漏检;其次,设计局部上下文增强(LCE)函数以更好地关注不同类别的遥感目标所需的上下文信息,从而避免误检;最后,使用加权双向特征金字塔网络(BiFPN)强化特征融合模块对遥感图像小目标的多尺度特征融合能力,从而改善算法检测效果。在DOTA数据集和NWPU VHR-10数据集上进行对比实验和消融实验,以验证所提算法的有效性和泛化性。实验结果表明,在2个数据集上所提算法的平均精确率均值(mAP)分别达到了77.33%和96.12%,而相较于YOLOv5算法,召回率分别提升了10.00和7.50个百分点。可见,所提算法能有效提升mAP和召回率,减少误检和漏检。

一种基于多状态颜色一致性约束的激光-惯性-视觉里程计

基于视觉、激光等单一传感器的定位方法难以适应多样化的环境,围绕激光雷达、惯性测量单元和相机3种模态的传感器信息源,针对激光雷达(Light Detection and Ranging,LiDAR)与视觉测量没有充分关联的问题,提出了一种基于多状态颜色一致性约束的激光雷达-惯性-视觉里程计方法,以提高系统的鲁棒性和定位精度。该方法紧耦合了激光雷达-惯性里程计(LiDAR-Inertial Odometry,LIO)和视觉-惯性里程计(Visual-Inertial Odometry,VIO)两个子系统,并定义了带有颜色信息的全局地图表示形式。LIO子系统中点云经过运动补偿后,直接用于构建点到面的残差。VIO子系统利用全局地图中点的深度信息,根据滑动窗口中多个相机状态观测到同一地图点颜色的一致性,构建光度误差约束,并通过不变扩展卡尔曼滤波(Extended Kalman Filter,EKF)状态估计器进行系统状态更新。在南洋理工大学发布的公共数据集上进行了实验,所提方法在该数据集不同序列上的绝对轨迹误差平均值为0.402 m。

基于反向最近邻的密度估计聚类算法

基于相互最近邻的密度峰聚类算法(DenMune)通过相互最近邻计算数据点的局部密度,是一种有效的聚类手段。但该算法存在构建聚类骨架不合理的问题,在分配弱点时采用硬投票策略,易产生错误。因此提出一种新的基于反向最近邻的密度估计聚类算法(RNN-DEC)。该算法引入反向最近邻来计算数据点的局部密度,将数据点分成强点、弱点和噪声点。使用强点构建聚类算法的骨架,通过软投票的方式将弱点分配到与其相似度最高的簇中去。提出了一种基于反向最近邻的簇融合算法,将相似度高的子簇融合,得到最终的聚类结果。实验结果表明,在一些合成数据集和UCI真实数据集上,相比较于其他经典算法,该算法具有更好的聚类效果。

基于特征融合的局部表征学习的遮挡行人重识别

为了增强行人重识别模型(Re-identification,ReID)的遮挡感知和局部特征捕捉能力,提出一种基于特征融合的局部表征学习的方法。首先,设计遮挡样本扩充策略,通过模拟多样化的遮挡场景,有效提升模型的鲁棒性和遮挡感知能力。其次,引入局部层次编码器,在全局语义的指导下提取序列的空间相关性特征,从而增强局部特征的可鉴别性和语义完整性。实验结果显示,在Occluded-Duke和Market-1501数据集上,该方法表现出色,特别是在Occluded-Duke数据集上的rank-1达到69.2%,优于现有先进方法,提升幅度为1.3百分点,验证了该方法在提升行人重识别任务性能方面的有效性。

新时代高校劳动教育与资助育人融合机制探赜——以商丘师范学院为例

劳动教育在国民教育体系中占据着举足轻重的地位,它是促进学生健康成长不可或缺的一环,具有塑造品德、增进智慧、强健体魄的综合育人价值。文章深入剖析地方高校商丘师范学院在融合劳动教育与资助育人方面的实践经验与问题,并提出一系列解决方案和建议,以促进这种融合机制的赓续发展,构建“以学习者为中心”的教育生态。这对于探索地方高校教育改革、促进学生综合素质提升,都具有一定的理论和实践意义。

扩展统一局部二值模式及图像纹理特征提取

统一局部二进制模式(ULBP)及其许多变体,已显示出对纹理分类的有效性。然而,这些ULBP方法中的大多数专注于编码中心像素与其相邻像素间的统一模式特征。因此存在无法捕捉非统一模式发生的像素间图像特征,忽略了不同邻域半径下像素间的相互作用这两个主要问题。针对于此,本文提出了扩展统一局部二值模式(EULBP)。EULBP统计图像相邻像素间的非统一模式,并逐步编码相邻采样点间的非统一模式;其次,在不同邻域下提取图像特征进行融合;最后,使用直方图交叉距离计算特征向量的相似度,得到在不同数据集中的识别率。经实验,验证了该算法的有效性。

Fusion框架系统的局部框架扰动的稳定性

该文在Hilbert空间中一般的框架序列扰动形式下,利用正交投影的性质和对偶框架的性质研究了原序列张成的闭子空间与扰动序列张成的闭子空间的关系,并探讨了局部框架的一般扰动对fusion框架系统稳定性的影响.这些结果推广和改进了由Casazza,Kutyniok和Li等得到的著名结果.

Hierarchical particle filter tracking algorithm based on multi-feature fusion

A hierarchical particle filter（HPF） framework based on multi-feature fusion is proposed.The proposed HPF effectively uses different feature information to avoid the tracking failure based on the single feature in a complicated environment.In this approach,the Harris algorithm is introduced to detect the corner points of the object,and the corner matching algorithm based on singular value decomposition is used to compute the firstorder weights and make particles centralize in the high likelihood area.Then the local binary pattern（LBP） operator is used to build the observation model of the target based on the color and texture features,by which the second-order weights of particles and the accurate location of the target can be obtained.Moreover,a backstepping controller is proposed to complete the whole tracking system.Simulations and experiments are carried out,and the results show that the HPF algorithm with the backstepping controller achieves stable and accurate tracking with good robustness in complex environments.

Multi-Behavior Fusion Based Potential Field Method for Path Planning of Unmanned Surface Vessel

The problem of the unmanned surface vessel (USV) path planning in static and dynamic obstacle environments is addressed in this paper. Multi-behavior fusion based potential field method is proposed, which contains three behaviors: goal-seeking, boundary-memory following and dynamic-obstacle avoidance. Then, different activation conditions are designed to determine the current behavior. Meanwhile, information on the positions, velocities and the equation of motion for obstacles are detected and calculated by sensor data. Besides, memory information is introduced into the boundary following behavior to enhance cognition capability for the obstacles, and avoid local minima problem caused by the potential field method. Finally, the results of theoretical analysis and simulation show that the collision-free path can be generated for USV within different obstacle environments, and further validated the performance and effectiveness of the presented strategy.

Medical image fusion based on pulse coupled neural networks and multi-feature fuzzy clustering

Medical image fusion plays an important role in clinical applications such as image-guided surgery, image-guided radiotherapy, noninvasive diagnosis, and treatment planning. In order to retain useful information and get more reliable results, a novel medical image fusion algorithm based on pulse coupled neural networks (PCNN) and multi-feature fuzzy clustering is proposed, which makes use of the multi-feature of image and combines the advantages of the local entropy and variance of local entropy based PCNN. The results of experiments indicate that the proposed image fusion method can better preserve the image details and robustness and significantly improve the image visual effect than the other fusion methods with less information distortion.

A MAP Approach for Vision-based Self-localization of Mobile Robot

一当场，自我本地化系统为在有深入的 3D 里程碑的 3D 环境起作用的活动机器人被开发。机器人通过合并从 odometry 和单向性的照相机收集的信息的一个地图评估者递归地估计它的姿势。我们为这二个传感器造非线性的模型并且坚持说机器人运动和不精密的传感器大小的无常操作应该全部被嵌入并且追踪我们的系统。我们在一个概率的几何学观点和使用 unscented 变换描述无常框架宣传无常，它经历给定的非线性的功能。就我们的机器人的处理力量而言，图象特征在相应投射特征的附近被提取。另外，数据协会被统计距离评估。最后，一系列系统的实验被进行证明我们的系统的可靠、精确的性能。

Subcellular Localization of Small GTP-binding Protein DsRab in Dunaliella salina

With the total RNA of Dunaliella salina as a template,the cD NA sequence of D. salina small GTP-binding protein gene was amplified by RT-PCR technique,and cloned onto pM Dl8-T simple vector,the recon was subjected to PCR detection and restriction endonuclease analysis,and the total sequence of DNA was determined. The results showed that the cloned fragment was 612 bp,and shared 100% homology with reported D. salina DsRab gene（ GenB ank： JN989548）. The target gene fragment was inserted downstream of pM DCG 35 S promoter,constructing subcellular localization recombinant vector pM DCG-DsRab. The successfully constructed subcellular localization recombinant vector pM DCG-DsRab was transformed into Agrobacterium tumefaciens LBA4404,and positive single clones were screened and used for transinfection of onion epidemical cells by Agrobacterium-mediated method,and the instant expression of DsRab was observed under fluorescence microscope. The results showed that the fusion protein GFP-DsRab was successfully expressed in onion epidemical cells,and mainly distributed on cytomembrane. This study will provide reference for further illumination of the function and action mechanism of D. salina small GTP-binding protein DsRab.

基于三维正态分布变换改进算法的移动机器人实时定位

针对点云配准三维正态分布变换(3D-NDT)在未确定初始位姿情况下配准精度较低、配准时间较长,无法满足移动机器人实时定位要求,提出1种改进的3D-NDT点云配准融合算法。在原始点云的降采样过程中,使用源点云中的点替代计算得到的重心,降低算法运算量并保留点云的特征信息;通过引入信赖半径动态调节迭代步长,提高降采样后的精度和点云配准速度;通过融合三维激光点云数据与9轴惯性测量单元(IMU)数据,解决2组点云数据位姿差异过大无法收敛或进入局部极值的问题。采用实验室自搭建的移动机器人平台对改进的3D-NDT算法进行仿真实验,验证改进算法实时定位的可靠性和准确度。结果表明:与传统3D-NDT算法相比,改进3D-NDT算法在室外和室内环境下的匹配精度分别提升106%,108%,匹配成功率分别提升8.29%,6.35%,平均匹配耗时分别降低51.1%,47.9%,移动机器人实时定位的配准精度得到较大提升,单次配准时间也大幅降低,改进的3D-NDT算法可满足移动机器人实时定位的需求。

RANDOM FOREST FOR INTERMEDIATE DESCRIPTOR FUSION IN SHOT BOUNDARY DETECTION

Shot boundary detection is the fundamental part in many real applications as video retrieval and so on. This paper tackles the problem of video segment obtaining in complex movie videos. Firstly, intermediate descriptor is proposed to depict the variation of both abrupt and gradual change in shot boundaries, which is formed by distance vector on Local Binary Pattern(LBP), GIST(GIST) or their fusion. Instead of just using the adjacent frames distance, intermediate descriptor keeps the distances between current frame and consecutive frames. It comprehensively characterizes local temporal structure, which is especially important for gradual change. For the excellent ability for feature fusion in random forests, it is adopted here to verify the fusion effect of intermediate descriptor on LBP and GIST. The whole experiments are designed on the subset of TRECVid 2013 INS(INstance Search) task to verify the effectiveness of proposed intermediate descriptor and the fusion ability for random forest. Compared with static and adaptive thresholds approaches, the best performance can be achieved by post-fusion of intermediate descriptor on LBP and GIST.