The Electromagnetic Scattering by Buried Complex Objects

The solution of scattering problem of buried complex bodies needs to not only consider the scattering from the complex bodies, but also apply the boundary conditions in two different types of coordinate systems. This paper applies the conversion technique of vector wave functions combined with the three-dimensional unimoment method to effectively solve this type of complex electromagnetic problem. Using the conversion relations between the cylindrical and spherical vector wave functions, two types of boundary-value problems are conveniently jointed. Both the vector wave function expansion and the multipole expansion belong to eigen function expansion methods and have the same convergence rate. But the former is more rigorous in theory and needs to calculate only two types of eigen components while the latter needs to calculate four types.

Concurrency Control Mechanism of Complex Objects

A complex object is an abstraction and description of a complex entity of the real world. Many applications in such domains as CIMS, CAD and OA define and manipulate a complex object as a single unit. In this paper, a definition of the model of complex objects is given, and the concurrency control mechanism of complex objects in WHYMX object-oriented database system is described.

A Deductive Database Approach for Complex Objects

A deductive database approach for complex objects reasoning is proposed,which is characterized by handling predicates nesting in terms of mapping hierarchically structured rules and facts to a flattened Horn-clause implementation scheme.

3D shape measurement of larger complex objects based on fringe cycle correction

The grating fringe on the reference plane is broadened in the intersecting axis system because of oblique-angle projection. In order to solve this problem, we study the theoretical model of the temporal phase unwrapping method based on the fringe cycle correction. We also study the 3D shape measurement theoretical model of the larger complex objects after considering the coordinate deviation and lens distortion. Experimental results demonstrate that the fringe cycle on the reference plane can be corrected to a constant value, the lens distortion can be corrected, and 3D shape of larger complex objects can be accurately measured.

Improved Multi-objective Ant Colony Optimization Algorithm and Its Application in Complex Reasoning

The problem of fault reasoning has aroused great concern in scientific and engineering fields.However,fault investigation and reasoning of complex system is not a simple reasoning decision-making problem.It has become a typical multi-constraint and multi-objective reticulate optimization decision-making problem under many influencing factors and constraints.So far,little research has been carried out in this field.This paper transforms the fault reasoning problem of complex system into a paths-searching problem starting from known symptoms to fault causes.Three optimization objectives are considered simultaneously: maximum probability of average fault,maximum average importance,and minimum average complexity of test.Under the constraints of both known symptoms and the causal relationship among different components,a multi-objective optimization mathematical model is set up,taking minimizing cost of fault reasoning as the target function.Since the problem is non-deterministic polynomial-hard(NP-hard),a modified multi-objective ant colony algorithm is proposed,in which a reachability matrix is set up to constrain the feasible search nodes of the ants and a new pseudo-random-proportional rule and a pheromone adjustment mechinism are constructed to balance conflicts between the optimization objectives.At last,a Pareto optimal set is acquired.Evaluation functions based on validity and tendency of reasoning paths are defined to optimize noninferior set,through which the final fault causes can be identified according to decision-making demands,thus realize fault reasoning of the multi-constraint and multi-objective complex system.Reasoning results demonstrate that the improved multi-objective ant colony optimization(IMACO) can realize reasoning and locating fault positions precisely by solving the multi-objective fault diagnosis model,which provides a new method to solve the problem of multi-constraint and multi-objective fault diagnosis and reasoning of complex system.

3D reconstruction of complex curved objects from line drawings

An active research topic in computer vision and graphics is developing algorithms that can reconstruct the 3D surface of curved objects from line drawings. There are a number of algorithms have been dedicated to solve this problem, but they can＇t solve this problem when the geometric structure of a curved object becomes complex. This paper proposes a novel approach to reconstructing a complex curved 3D object from single 2D line drawings. Our approach has three steps： （1） decomposing a complex line drawing into several simpler line drawings and transforming them into polyhedron; （2） reconstructing the 3D wireframe of curved object from these simpler line drawings and generating the curved faces; （3） combining the 3D objects into the complete objects. A number of examples are given to demonstrate the ability of our approach to successfully perform reconstruction of curved objects which are more complex than previous methods.

Tracking of Non-Rigid Object in Complex Wavelet Domain

In this paper we have proposed an object tracking method using Dual Tree Complex Wavelet Transform (DTCxWT). The proposed method is capable of tracking the moving object in video sequences. The object is assumed to be deform-able under limit i.e. it may change its shape from one frame to another. The basic idea in the proposed method is to decompose the image into two components: a two dimensional motion and a two dimensional shape change. The motion component is factored out while the shape is explicitly represented by storing a sequence of two dimensional models. Each model corresponds to each image frame. The proposed method performs well when the change in the shape in the consecutive frames is small however the 2-D motion in consecutive frames may be large. The proposed algorithm is capable of handling the partial as well as full occlusion of the object.

OBJECT ORIENTED DATA MODELLING WITH APPLICATIONS TO CAD

An object oriented data modelling in computer aided design (CAD) databases is focused. Starting with the discussion of data modelling requirements for CAD applications, appropriate data modelling features are introduced herewith. A feasible approach to select the “best” data model for an application is to analyze the data which has to be stored in the database. A data model is appropriate for modelling a given task if the information of the application environment can be easily mapped to the data model. Thus, the involved data are analyzed and then object oriented data model appropriate for CAD applications are derived. Based on the reviewed object oriented techniques applied in CAD, object oriented data modelling in CAD is addressed in details. At last 3D geometrical data models and implementation of their data model using the object oriented method are presented.

Thinking in three's: Changing surgical patient safety practices in the complex modern operating room

The three surgical patient safety events, wrong site surgery, retained surgical items (RSI) and surgical fires are rare occurrences and thus their effects on the complex modern operating room (OR) are difficult to study. The likelihood of occurrence and the magnitude of risk for each of these surgical safety events are undefined. Many providers may never have a personal experience with one of these events and training and education on these topics are sparse. These circumstances lead to faulty thinking that a provider won't ever have an event or if one does occur the provider will intuitively know what to do. Surgeons are not preoccupied with failure and tend to usually consider good outcomes, which leads them to ignore or diminish the importance of implementing and following simple safety practices. These circumstances contribute to the persistent low level occurrence of these three events and to the difficulty in generating sufficient interest to resource solutions. Individual facilities rarely have the time or talent to understand these events and develop lasting solutions. More often than not, even the most well meaning internal review results in a new line to a policy and some rigorous enforcement mandate. This approach routinely fails and is another reason why these problems are so persistent. Vigilance actions alone havebeen unsuccessful so hospitals now have to take a systematic approach to implementing safer processes and providing the resources for surgeons and other stake-holders to optimize the OR environment. This article discusses standardized processes of care for mitigation of injury or outright prevention of wrong site surgery, RSI and surgical fires in an action-oriented framework illustrating the strategic elements important in each event and focusing on the responsibilities for each of the three major OR agents-anesthesiologists, surgeons and nurses. A Surgical Patient Safety Checklist is discussed that incorporates the necessary elements to bring these team members together and influence the emergence of a safer OR.

Testability Models for Object-Oriented Frameworks

Frameworks are time-tested highly reusable architectural skeleton structures. They are designed ‘abstract’ and ‘inco- mplete’ and are designed with predefined points of variability, known as hot spots, to be customized later at the time of framework reuse. Frameworks are reusable entities thus demand stricter and rigorous testing in comparison to one- time use application. The overall cost of framework development may be reduced by designing frameworks with high testability. This paper aims at discussing a few metric models for testability analysis of object-oriented frameworks in an attempt to having quantitative data on testability to be used to plan and monitor framework testing activities so that the framework testing effort and hence the overall framework development effort may be brought down.

基于二阶段目标增强网络的低照度复杂环境下绝缘子故障检测方法

从低照度户外环境中航拍采集的绝缘子影像,存在照度低、背景复杂、绝缘子故障目标小等缺陷,严重影响低照度环境下绝缘子故障检测准确性。为解决上述问题,文章提出一种基于TOE-Net的低照度复杂环境下绝缘子故障检测方法,提出TOE-Net进行图像预处理方法,再使用YOLOv7-OL作为故障检测模块检测小目标绝缘子故障。在二阶段目标增强网络(two-stage object enhancement network,TOE-Net)中,设计零目标图像增强损失函数实现预增强网络(preparation enhancement network,PreEnNet)和深度增强网络(deep enhancement network,DeepEnNet)的无监督学习;使用信道级注意力模块跳跃式通道注意力机制(skip squeeze excitation networt,Skip_SENet)和跳跃式通道注意力机制(skip convolutional block attention module,Skip_CBAM)模块改进原始小目标特征增强单次多框检测算法(small object detection enhancement single shot multiBox detector,SDE-SSD),从而提升定位网络的小目标检测能力;设计弱监督机制使预增强网络根据小目标特征增强SSD的要求来提升图像增强能力,直到小目标特征增强SSD能够从增强图像中准确定位绝缘子串位置;使用深度增强网络深度增强绝缘子串区域,提升各类故障的特征显著性。故障检测模块中,将YOLOv7目标检测算法改进为面向小目标YOLOv7,在原模型中添加结合多尺度特征自适应融合网络的小目标检测通道,并将原始损失函数的CIOU改进为BIOU,从而提高模型的小目标检测性能。在低照度环境绝缘子故障检测实验中,该算法与5种目前常用目标检测算法相比具有较大优势,并且相较于低光目标检测算法IA-YOLO、GenISP with RetinaNet,m AP提升9.77%、10.35%,检测速度提升7.23%、10.16%,证明该算法适用于低照度复杂环境下小目标绝缘子故障检测任务;在正常光照绝缘子故障检测实验中该算法仍保持出色性能,证明该算法能够实现常规光照条件下绝缘子小目标故障检测。

基于GLMB滤波的复杂场景下红外弱小目标自适应跟踪算法

针对红外弱小目标在复杂场景下受到漏检和杂波影响,导致跟踪不连续甚至失效的问题,本文提出一种红外弱小目标自适应跟踪算法。在预处理阶段,为了减少不必要的计算,首先定义一种衡量图像复杂度的算法。然后该算法通过计算红外图像多个特征得到场景复杂度来确认场景类型,再根据场景类型选取对应的检测算法提取目标候选位置、灰度以及局部直方图等特征建立对应的量测模型与似然函数。在目标跟踪阶段,为了自适应地匹配广义标签多伯努利(Generalized Labeled Multi-Bernoulli,GLMB)滤波器的滤波参数,在GLMB的基础上提出一种适应视频图像的新生算法进行航迹起始;针对红外图像序列目标检测概率未知的情况,将未知检测概率的基数化概率假设密度(Cardinality Probability Hypothesis Density,CPHD)滤波器集成到GLMB中实时估计目标检测概率以提升跟踪精度。仿真结果表明,所提出算法能有效地排除量测漏检和虚警的干扰,跟踪不同红外复杂场景下的弱小目标。

基于YOLOv5的复杂背景下植物叶片检测研究

对植物叶片进行检测是研究植物表型性状的基础,但真实环境下叶片间相互遮挡、叶片边缘特征不明显、幼叶目标过小以及外部环境如光照条件等因素影响会对叶片检测效果造成很大的障碍。针对复杂背景下的叶片检测,该研究提出了一种基于改进YOLOv5模型植物叶片检测方法。通过在骨干网络中引入空洞卷积,使得网络可以捕获到更广阔范围的上下文信息;利用双向连接的加权特征金字塔网络,以增强目标叶片特征提取并更好地融合特征信息;利用注意力机制,通过动态地调整注意力分布,以提高边缘特征表达能力。测试结果表明,在Plant Village数据集筛选的葡萄叶片图像以及自拍摄葡萄生长叶片上测试改进算法的可行性,改进的YOLOv5模型其叶片检测mAP比原生模型提高了5.8%,遮挡叶片检测精度提高了7.09%。叶片检测效果有显著提升。该研究提出的方法可以有效解决复杂背景下植物叶片检测效果不佳的问题,为植物表型研究提供技术支撑。

基于改进YOLO v5的复杂环境下柑橘目标精准检测与定位方法

针对自然环境下柑橘果实机械化采收作业环境复杂和果实状态多样等情况,提出了一种多通道信息融合网络——YOLO v5-citrus,以解决柑橘果实识别精准度低、果实分类模糊和定位精准度低等难题。将不同的柑橘目标通过不同遮挡条件分为“可采摘”和“难采摘”两类,这种分类策略可指导机器人在真实果园中顺序摘取,提高采摘效率并减少机器人本体和末端执行器损坏率。YOLO v5-citrus中,在颈部网络插入多通道信息融合模块,对柑橘的深浅特征信息进行处理,提高柑橘采摘状态识别精度,同时修改颈部网络拼接方法,针对目标柑橘大小进行识别,训练后在识别部分嵌入聚类算法模块,将训练部分识别模糊的柑橘目标进行最后区分。识别后进行深度图像和彩色图像的像素对齐,并通过坐标系转换获取柑橘目标三维坐标。在使用多种增强技术处理的数据集中,YOLO v5-citrus比原始YOLO v5在平均精度均值和精确率上分别提高2.8个百分点与3.7个百分点,表现出更优异的泛化能力。与YOLO v7和YOLO v8等其他主流网络架构相比较,保持了更高的检测精度和更快的检测速度。通过真实果园的检测与定位试验,得到柑橘目标的三维坐标识别定位系统的定位误差为(1.97 mm,0.36 mm,9.63 mm),满足末端执行器的抓取条件。试验结果表明,该模型具有较强的鲁棒性,满足复杂环境下柑橘状态识别要求,可为柑橘园机械采收设备提供技术支持。

面向复杂田间收获作业的轻量化三七目标检测方法

针对目前三七检测算法在复杂田间收获工况下检测精度低、模型复杂度大、移动端部署难等问题,该研究提出一种基于YOLOv5s的轻量化三七目标检测方法。首先,采用GSConv卷积方法替换原始颈部网络的传统卷积,引入Slim-neck轻量级颈部网络,降低了模型复杂度,同时提升了模型精度;其次,使用ShuffleNetv2轻量型特征提取网络对主干网络进行轻量化改进,提升了模型实时检测性能,并采用角度惩罚度量的损失(SIoU)优化边界框损失函数,提升了轻量化后的模型精度和泛化能力。试验结果表明,改进后的PN-YOLOv5s模型参数量、计算量、模型大小分别为原YOLOv5s模型的46.65%、34.18%和48.75%,检测速度提升了1.2倍,F_(1)值较原始模型提升了0.22个百分点,平均精度均值达到了94.20%,较原始模型低0.6个百分点,与SSD、Faster R-CNN、YOLOv4-tiny、YOLOv7-tiny和YOLOv8s模型相比能够更好地平衡检测精度与速度,检测效果更好。台架试验测试结果表明,4种输送分离作业工况下三七目标检测的准确率达90%以上,F_(1)值达86%以上,平均精度均值达87%以上,最低检测速度为105帧/s,实际收获工况下模型的检测性能良好,可为后续三七收获作业质量实时监测与精准分级输送提供技术支撑。

融合卷积和注意力机制的车辆行人检测算法

针对复杂场景下的车辆、行人检测困难的问题,提出一种融合卷积和注意力机制的车辆行人检测算法ATDWYOLOX(ATtention DepthWise YOLOX)。在主干网络中采用CSPNeXt,并设计融合卷积和注意力机制的模块ACMBottleneck(Attention Convolution Mix Bottleneck);在特征融合网络中,设计融合深度可分离卷积和注意力机制的网络ATDW-PAFPN(ATtention DepthWise Progressive Asymmetric Feature Pyramid Network);在损失函数中,使用完全交并比(CIoU)和Varifocal Loss。在公开数据集BDD100K上的实验结果表明,所提算法平均检测准确率(mAP)相较于YOLOX提升了2.52个百分点,能较好地完成复杂场景下车辆行人的检测任务。

Cyclomatic Complexity-Based Encapsulation, Data Hiding, and Separation of Concerns

Three design principles are prominent in software development-encapsulation, data hiding, and separation of concerns. These principles are used as subjective quality criteria for both procedural and object-oriented applications. The purpose of research is to quantify encapsulation, data hiding, and separation of concerns is quantified using cyclomatic-based metrics. As a result of this research, the derived design metrics, coefficient of encapsulation, coefficient of data hiding, and coefficient of separation of concerns, are defined and applied to production software indicating whether the software has low or high encapsulation, data hiding, and separation of concerns.

CSAM:用于单目标跟踪的卷积自注意力模块

为了在目标发生遮挡、形变、尺度变化和背景干扰等场景下准确地跟踪目标,注意力机制被广泛应用于特征抽取模块,以选择性地关注重要特征和抑制无关特征.然而,现有的注意力机制只考虑了通道特征层与空间特征点之间的局部或全局关系,没有对特征进行融合建模.本文针对复杂跟踪场景提出了一种基于卷积神经网络和自注意力机制的卷积自注意力模块(Convolutional Self-Attention Module,CSAM),该模块能够以注意力加权方式解决前景遮蔽、非刚性形变、快速尺度变化与相似特征背景干扰问题.经过实验验证,引入卷积自注意力模块的孪生网络能够显著地提升跟踪器的性能,在跟踪问题基准(Benchmark)数据集OTB100上以平均重叠率、跟踪成功率与准确率作为评判指标,相比基准模型分别提升了9.2%、2.2%与2.9%.通过进一步的消融实验证明了本文提出并引入的适用于孪生网络跟踪框架的卷积自注意力模块能够有效地提升特征辨识度,对比先进方案兼顾单目标跟踪性能和实时性,能够在大多数实时复杂跟踪场景实现轻量化部署.

基于l^(1/2)-TV正则化RPCA的运动目标检测

针对复杂环境下背景干扰导致运动目标检测精度下降的问题,提出了一种基于l^(1/2)-TV正则化RPCA的运动目标检测方法。方法利用核范数来描述背景的低秩特性,采用l^(1/2)范数描述更稀疏的运动目标,以抑制运动目标中的背景干扰。同时结合TV正则化约束运动目标的空间连续性,使运动目标更加完整。利用Frobenius范数检测背景干扰。采用交替方向最小化策略扩展的增广拉格朗日乘子法求解所提出的约束最小化问题。实验结果表明,所提方法能有效去除背景干扰,提高运动目标的检测精度、改善视觉效果。

基于改进YOLOv5的田间大豆花朵生长状态识别方法研究

为判断大豆开花时期落花情况,对田间大豆花朵在花蕾、半开、全开、凋落四类生长状态下进行精准检测。基于YOLOv5检测模型,对主干Bottleneck CSP结构进行修改,减少模块数量来保留更多浅层特征,增强特征表达能力,在骨干网络中引入CA注意机制,以获得位置信息,协助模型更加准确地识别,并修改锚箱尺寸提高小目标花蕾精准识别,在自建的田间大豆花朵不同生长状态数据集上进行改进YOLOv5算法对比试验。结果表明:大豆开花时期花朵不同生长状态识别模型准确率达到93.4%,召回率达到91.4%,对比原模型准确率、召回率分别提高0.8%和2.1%。