YOLO-MFD:Remote Sensing Image Object Detection with Multi-Scale Fusion Dynamic Head

Remote sensing imagery,due to its high altitude,presents inherent challenges characterized by multiple scales,limited target areas,and intricate backgrounds.These inherent traits often lead to increased miss and false detection rates when applying object recognition algorithms tailored for remote sensing imagery.Additionally,these complexities contribute to inaccuracies in target localization and hinder precise target categorization.This paper addresses these challenges by proposing a solution:The YOLO-MFD model(YOLO-MFD:Remote Sensing Image Object Detection withMulti-scale Fusion Dynamic Head).Before presenting our method,we delve into the prevalent issues faced in remote sensing imagery analysis.Specifically,we emphasize the struggles of existing object recognition algorithms in comprehensively capturing critical image features amidst varying scales and complex backgrounds.To resolve these issues,we introduce a novel approach.First,we propose the implementation of a lightweight multi-scale module called CEF.This module significantly improves the model’s ability to comprehensively capture important image features by merging multi-scale feature information.It effectively addresses the issues of missed detection and mistaken alarms that are common in remote sensing imagery.Second,an additional layer of small target detection heads is added,and a residual link is established with the higher-level feature extraction module in the backbone section.This allows the model to incorporate shallower information,significantly improving the accuracy of target localization in remotely sensed images.Finally,a dynamic head attentionmechanism is introduced.This allows themodel to exhibit greater flexibility and accuracy in recognizing shapes and targets of different sizes.Consequently,the precision of object detection is significantly improved.The trial results show that the YOLO-MFD model shows improvements of 6.3%,3.5%,and 2.5%over the original YOLOv8 model in Precision,map@0.5 and map@0.5:0.95,separately.These results illustrate the clear advantages of the method.

Dynamic-YOLOX:复杂背景下的苹果叶片病害检测模型

针对目前苹果叶片数据集的叶片病害种类不全以及图片背景单一等问题,构建了复杂背景下包括苹果叶部六种常见病害的苹果叶片病害数据集。针对目前主流苹果叶片病害检测模型检测精度不高、模型复杂和不满足实时监测等问题,提出了一种基于YOLOX-S(you only look once X-S)改进得到的复杂背景下的苹果叶片病害自适应检测模型Dynamic-YOLOX。设计并使用ECA-SPPFCSPC模块(efficient channel attention cross-stage partial fast spatial pyramid pooling module)更换YOLOX-S模型主干网络尾部Dark5中的空间金字塔池化(SPP)以及跨阶段局部网络(CSPNet)模块来增强模型关注深层语义特征、抑制无用信息的能力,并减少硬件内存开销。设计了动态跨阶段局部网络(ODCSP)模块,并用其更换YOLOX-S模型主干网络中Dark2、Dark3、Dark4部分以及颈部网络中所有的CSPNet模块,使得模型在面对不同输入特征时有更强的自适应性,在减少模型的参数量和计算量的同时提高了模型的平均检测精度。引入Varifocal Loss更换模型中分类置信度损失的BCEWithLogits Loss来提升模型对苹果叶片中密集小目标病害的检测精度。在自制数据集上Dynamic-YOLOX相对原始YOLOX-S模型的mAP提升了4.54个百分点,达到84.63%,同时模型的参数量和计算量分别下降了11.97%和13.45%,检测速度达到44.07 FPS。对比主流苹果叶片病害检测模型,Dynamic-YOLOX具有一定优越性。

MOVING OBJECT TRACKING IN DYNAMIC IMAGE SEQUENCE BASED ON ESTIMATION OF MOTION VECTORS OF FEATURE POINTS

An improved estimation of motion vectors of feature points is proposed for tracking moving objects of dynamic image sequence. Feature points are firstly extracted by the improved minimum intensity change （MIC） algorithm. The matching points of these feature points are then determined by adaptive rood pattern searching. Based on the random sample consensus （RANSAC） method, the background motion is finally compensated by the parameters of an affine transform of the background motion. With reasonable morphological filtering, the moving objects are completely extracted from the background, and then tracked accurately. Experimental results show that the improved method is successful on the motion background compensation and offers great promise in tracking moving objects of the dynamic image sequence.

Visual SLAM in dynamic environments based on object detection

A great number of visual simultaneous localization and mapping(VSLAM)systems need to assume static features in the environment.However,moving objects can vastly impair the performance of a VSLAM system which relies on the static-world assumption.To cope with this challenging topic,a real-time and robust VSLAM system based on ORB-SLAM2 for dynamic environments was proposed.To reduce the influence of dynamic content,we incorporate the deep-learning-based object detection method in the visual odometry,then the dynamic object probability model is added to raise the efficiency of object detection deep neural network and enhance the real-time performance of our system.Experiment with both on the TUM and KITTI benchmark dataset,as well as in a real-world environment,the results clarify that our method can significantly reduce the tracking error or drift,enhance the robustness,accuracy and stability of the VSLAM system in dynamic scenes.

A Dynamic Object Behavior Model and Implementation Based on Computational Reflection

A dynamic object behavior model based on computational reflection is proposed. This model consists of function level and meta level, the meta objects in meta level manage the base objects and behaviors in function level, including dynamic binding and unbinding of base object and behavior. We implement this model with RoleJava Language,which is our self linguistic extension of the Java Language.Meta Objects are generated automatically at compile\|time,this makes the reflecton mechanism transparent to programmers.Finally an example applying this model to a banking system is presented.

Adaptive key SURF feature extraction and application in unmanned vehicle dynamic object recognition

A new method based on adaptive Hessian matrix threshold of finding key SRUF （ speeded up robust features） features is proposed and is applied to an unmanned vehicle for its dynamic object recognition and guided navigation. First, the object recognition algorithm based on SURF feature matching for unmanned vehicle guided navigation is introduced. Then, the standard local invariant feature extraction algorithm SRUF is analyzed, the Hessian Metrix is especially discussed, and a method of adaptive Hessian threshold is proposed which is based on correct matching point pairs threshold feedback under a close loop frame. At last, different dynamic object recognition experi- ments under different weather light conditions are discussed. The experimental result shows that the key SURF feature abstract algorithm and the dynamic object recognition method can be used for un- manned vehicle systems.

Secure Synchronization Control for a Class of Cyber-Physical Systems With Unknown Dynamics

This paper investigates the secure synchronization control problem for a class of cyber-physical systems(CPSs)with unknown system matrices and intermittent denial-of-service(DoS)attacks.For the attack free case,an optimal control law consisting of a feedback control and a compensated feedforward control is proposed to achieve the synchronization,and the feedback control gain matrix is learned by iteratively solving an algebraic Riccati equation(ARE).For considering the attack cases,it is difficult to perform the stability analysis of the synchronization errors by using the existing Lyapunov function method due to the presence of unknown system matrices.In order to overcome this difficulty,a matrix polynomial replacement method is given and it is shown that,the proposed optimal control law can still guarantee the asymptotical convergence of synchronization errors if two inequality conditions related with the DoS attacks hold.Finally,two examples are given to illustrate the effectiveness of the proposed approaches.

Key techniques for predicting the uncertain trajectories of moving objects with dynamic environment awareness

Emerging technologies of wireless and mobile communication enable people to accumulate a large volume of time-stamped locations,which appear in the form of a continuous moving object trajectory.How to accurately predict the uncertain mobility of objects becomes an important and challenging problem.Existing algorithms for trajectory prediction in moving objects databases mainly focus on identifying frequent trajectory patterns,and do not take account of the effect of essential dynamic environmental factors.In this study,a general schema for predicting uncertain trajectories of moving objects with dynamic environment awareness is presented,and the key techniques in trajectory prediction arc addressed in detail.In order to accurately predict the trajectories,a trajectory prediction algorithm based on continuous time Bayesian networks（CTBNs） is improved and applied,which takes dynamic environmental factors into full consideration.Experiments conducted on synthetic trajectory data verify the effectiveness of the improved algorithm,which also guarantees the time performance as well.

Mobile Internet Mobile Agent System Dynamic Trust Model for Cloud Computing

In mobile cloud computing,trust is a very important parameter in mobile cloud computing security because data storage and data processing are performed remotely in the cloud.Aiming at the security and trust management of mobile agent system in mobile cloud computing environment,the Human Trust Mechanism(HTM)is used to study the subjective trust formation,trust propagation and trust evolution law,and the subjective trust dynamic management algorithm(MASTM)is proposed.Based on the interaction experience between the mobile agent and the execution host and the third-party recommendation information to collect the basic trust data,the public trust host selection algorithm is given.The isolated malicious host algorithm and the integrated trust degree calculation algorithm realize the function of selecting the trusted cluster and isolating the malicious host,so as to enhance the security interaction between the mobile agent and the host.Given algorithm simulation and verification were carried out to prove its feasibility and effectiveness.

Research on modeling and applying system of combination object

While studying Descriptive Geometry & Mechanical Drawing, students always feel that it is very difficult to develop space imagining ability. This paper researches the modeling and applying system of combination object with high practicality. Through dynamic displays, students can observe the outlines of combination object vividly and roundly. At the same time, they can also quickly develop their space imagining abilities. The whole system is developed on the personal computer with VC++5.0, based on 3D MAX, MFC and OpenGL of PC.

Dynamic Slicing of Object Oriented Programs

Program slice has many applications such as program debugging, testing, maintenance, and complexity measurement. A static slice consists of all statements in program P that may effect the value of variable v at some point p , and a dynamic slice consists only of statements that influence the value of variable occurrence for specific program inputs. In this paper, we concern the problem of dynamic slicing of object oriented programs which, to our knowledge, has not been addressed in the literatures. To solve this problem, we present the dynamic object oriented dependence graph (DODG)which is an arc classified digraph to explicitly represent various dynamic dependence between statement instances for a particular execution of an object oriented program. Based on the DODG, we present a two phase backward algorithm for computing a dynamic slice of an object oriented program.

Security and Stability Aspects of Multi Objective Dynamic Economic Dispatch with Renewable Energy and HVDC Transmission Lines

Renewable sources of energy are being integrated into the power grids due to their economic and environmental merits as compared with the traditional fossil-fuel-fired power generation. However, their significant penetration demands a thorough research in terms of system reliability, that is, security and stability. In this paper, Security Constrained Multi Objective Dynamic Economic Dispatch (SCMODED) problem considering cubic thermal cubic cost function, wind, solar penetration, cubic transmission power losses and cubic emissions cost function as objectives is first formulated. Both HVDC and HVAC lines are included in their formulation. Various approaches like probabilistic load flow (PLF), scenario based method, participation factors and Harmony Search algorithm etc. are employed in the solution process. Security and stability effects of renewable energy (RE) penetration are investigated and analyzed. The simulated results reveal that RE penetration leads to reduced cost and emissions and increased security concerns. Further, there is increased power system instability and hence increased load shedding so as to help the power system attain steady state stability. Inclusion of HVDC lines facilitates rapid and fast control to increase the transient stability limit by the action of the converter ignition angle (CIA) and converter extinction angle (CEA).

Immune Optimization Approach for Dynamic Constrained Multi-Objective Multimodal Optimization Problems

This work investigates one immune optimization approach for dynamic constrained multi-objective multimodal optimization in terms of biological immune inspirations and the concept of constraint dominance. Such approach includes mainly three functional modules, environmental detection, population initialization and immune evolution. The first, inspired by the function of immune surveillance, is designed to detect the change of such kind of problem and to decide the type of a new environment;the second generates an initial population for the current environment, relying upon the result of detection;the last evolves two sub-populations along multiple directions and searches those excellent and diverse candidates. Experimental results show that the proposed approach can adaptively track the environmental change and effectively find the global Pareto-optimal front in each environment.

Object-oriented GIS technology and mine geo-dynamic division information management

Object-oriented methodology was expected to revolutionize GIS design and development, making it easier and faster to design, through a common approach to conceptual modelling, database design, programming languages and user interface tools. In addition, an object-oriented model provides some useful tools for data abstraction and data structuring, which augment the conventional tools and overcome some deficiencies inherent to the traditional relational model. The principles of object-oriented modelling applied to geographic data were put forward, and the concepts of object-oriented paradigm and characteristics of spatial data management were proposed in order to argument their compatibility.

Dynamic Multi-objective Optimization of Chemical Processes Using Modified BareBones MOPSO Algorithm

Dynamic multi-objective optimization is a complex and difficult research topic of process systems engineering. In this paper,a modified multi-objective bare-bones particle swarm optimization( MOBBPSO) algorithm is proposed that takes advantage of a few parameters of bare-bones algorithm. To avoid premature convergence,Gaussian mutation is introduced; and an adaptive sampling distribution strategy is also used to improve the exploratory capability. Moreover, a circular crowded sorting approach is adopted to improve the uniformity of the population distribution.Finally, by combining the algorithm with control vector parameterization,an approach is proposed to solve the dynamic optimization problems of chemical processes. It is proved that the new algorithm performs better compared with other classic multiobjective optimization algorithms through the results of solving three dynamic optimization problems.

基于AMOWOA的区域综合能源系统运行优化调度

目前,智能优化算法已广泛应用于工程优化中,在当前多能耦合与互补的能源发展趋势下,仅考虑系统经济指标的单目标优化模式已经不再适用于目前区域综合能源系统(Integrated energy system, IES)的运行优化调度,需要研究一种多目标运行策略来解决区域综合能源系统的运行优化调度问题.首先综合考虑经济与能源利用两个指标并结合商业住宅区域的特性,以系统日运行收益和一次能源利用率为优化目标构建商业住宅区域综合能源系统多目标运行优化调度模型.其次由于传统多目标智能优化算法缺乏一种最优解综合评价方法,基于非支配排序以及拥挤度计算的多目标算法框架,提出一种利用模糊一致矩阵选取全局最优解的多目标鲸鱼优化算法(A multi-objective whale optimization algorithm, AMOWOA),并将提出算法对商住区域综合能源系统多目标运行优化调度模型进行求解.最后以华东某商业住宅区域综合能源系统为例进行仿真,验证了该方法的有效性和可行性.

基于数据流的K-S变化检测的动态多目标规划算法

为了更加准确地判断环境是否发生变化并快速追踪动态多目标规划问题(dynamicmulti-objectiveoptimization problem,DMOP)当前时刻的Pareto前沿,提出了一种基于数据流的Kolmogorov-Smirnov(K-S)变化检测的动态多目标规划(DSK-SDMOP)算法。该算法以NSGA-Ⅱ为基础,通过数据流建立2个时刻的检验窗口,再利用K-S检验基于数据流的Pareto最优前沿是否发生变化,检测2个窗口的数据是否服从同一分布来判断环境是否发生变化,并就环境变化的剧烈程度实行相应的应答机制,以提高对环境的适应程度。利用基于数据流的K-S检测方法,对环境变化不会过于敏感,而且不用提前假设对应目标值的分布,易于操作。通过5个动态多目标规划标准测试函数对该算法进行测试,并和现有的2种算法进行对比分析,结果表明该算法处理动态多目标规划问题具有良好的性能。

基于多目标灰狼算法的漂浮式风电机组浮台内TMD参数优化

针对漂浮式风电机组浮台内调谐质量阻尼器(TMD)参数调优的问题,以5 MW Barge型漂浮式风电机组为研究对象,采用多目标灰狼算法(MOGWO)优化TMD参数配置。首先,基于欧拉-拉格朗日方程建立浮台内含TMD的漂浮式风电机组动力学模型,采用Levenberg-Marquardt(LM)法进行模型未知参数辨识;其次,同时考虑塔顶和塔基控制目标,采用MOGWO算法优化TMD的刚度和阻尼参数;最后,在不同工况下进行仿真分析。结果表明:相对于传统的单目标优化算法,使用MOGWO算法参数优化后的TMD对风电机组具有更好的振动抑制效果。

改进动态规划算法求解同尺寸物品的装箱问题

装箱问题是在一个容量有限的箱内尽可能多的装入各类物品。文中研究同尺寸的物品装箱问题,在一个集装箱中装入大小规格一致的小箱,使集装箱的空间利用率最大,即装入的小箱数量最多。采用分层装载思想和同质条带的布局方式设计装箱方案,利用改进的动态规划算法求解层装的布局问题,使用背包算法求解同质条带内的小箱布局问题。与已有文献算法相比,文中算法的运行速度更快,得到的装箱方案更便捷,利于装箱操作,而且在一定程序上提高了装箱率。使用国际标准尺寸的集装箱和随机尺寸的物品小箱进行实验,结果表明文中算法比传统的动态规划算法有更好的求解效果,可适用于实际的集装箱物品装载。

改进多目标蜂群算法优化洗出运动及仿真实验

针对经典洗出算法参数选择不当导致信号缺失,引起失真,影响洗出效果等问题,提出一种改进的多目标人工蜂群算法,利用该算法对经典洗出算法中的滤波器参数进行优化来改善洗出算法的洗出效果。针对传统蜂群算法初始化和局部优化中存在的问题,引入Circle映射和Pareto局部优化算法;建立人体感知误差模型、加速度差值模型、位移模型,将模型函数作为目标函数,用改进后的多目标人工蜂群算法对经典洗出算法进行参数优化;建立仿真模型对优化后的洗出算法进行仿真验证,应用飞行模拟器运动实验平台进行实验验证。结果表明:经优化后的洗出算法,洗出逼真度得到有效提升,降低了误差峰值,改善了相位延迟,节省了运动空间。