相机标定是视觉测量中的关键步骤,其目的是确定相机的内外参数,以提高图像处理的精度和可靠性。为解决传统相机标定中标定精度低、鲁棒性弱等一系列问题,提出了基于鲸鱼优化算法的相机标定方法,本方法结合鲸鱼优化算法的全局搜索能力,优化相机参数。在实验中,首先通过张正友标定法获取相机的初始参数。随后,使用鲸鱼优化算法在全局范围内对标定参数进行优化,以最小化重投影误差。实验结果表明,基于鲸鱼优化算法的标定方法相比传统方法具有更高的精度和更强的鲁棒性。Camera calibration is a critical step in vision measurement, the purpose of which is to determine the internal and external parameters of the camera to improve the accuracy and reliability of image processing. In order to solve a series of problems such as low calibration accuracy and weak robustness in traditional camera calibration, a camera calibration method based on whale optimization algorithm was proposed, which combined with the global search ability of whale optimization algorithm to optimize camera parameters. In the experiment, the initial parameters of the camera were obtained by Zhang Zhengyou’s calibration method. Subsequently, the whale optimization algorithm was used to optimize the calibrated parameters on a global scale to minimize the reprojection error. Experimental results show that the calibration method based on whale optimization algorithm has higher accuracy and stronger robustness than the traditional method.
玉米作为全球最重要的粮食作物之一,其产量预测在确保粮食安全、优化农业管理以及支持政策制定方面具有重要意义。本文系统综述了当前玉米产量预测领域的主要方法,包括传统的作物预测系统模型、数据驱动的机器学习模型以及近年来迅速发展的深度学习模型,深入分析了气候、土壤、遥感和社会经济等多种预测变量对产量预测精度的影响。通过对现有研究的成果进行总结,本文揭示了不同模型的优劣势,并指出了现阶段研究中存在的不足,如模型泛化能力有限、多模态数据融合的挑战等。最后,本文对未来研究提出了展望,建议进一步探索多源数据的整合、优化模型的适应性以及提升预测的解释性,以推动玉米产量预测技术向更精准、智能化方向发展。本文为未来的玉米产量预测研究提供了有价值的理论基础和实践指导。Maize is one of the most important crops globally, and accurate yield prediction is crucial for food security, agricultural management, and policy decisions. This paper reviews key methods for maize yield prediction, including crop prediction models, machine learning, and deep learning. It analyzes the impact of variables like climate, soil, remote sensing, and socioeconomic factors on prediction accuracy. By summarizing current research, this paper highlights the strengths and limitations of these models and addresses challenges such as limited generalization and multimodal data integration. Finally, it proposes future directions for improving model adaptability and prediction precision, offering valuable insights for further research.
空气质量一直是公众和社会最关注的环境问题之一,其质量已经因为经济活动而受到损害。然而,由于空气的复杂性以及人类活动对空气的影响,预测空气质量是一项不小的挑战。传统上使用单一的机器学习模型预测空气质量时,在各种AQI波动趋势下难以获得良好的预测结果。为了有效地解决这一问题,本文提出了一种分层置信规则库(Hierarchy Belief Rule Base-RFI)预测模型,为空气质量预测提供了一种有用的方法。首先,为了提高模型的精准度,采用了随机森林来对关键特征进行筛选,同时对未被筛选出的特征进行ER融合。其次,提出了空气质量预测方法的可解释性准则,以规范整个建模过程的可解释性。然后给出了模型的推理和优化方法。最后通过实验验证了该模型的有效性。Air quality has always been one of the top environmental concerns of the public and society, and its quality has been compromised by economic activities. However, due to the complexity of the air and the impact of human activities on it, predicting air quality is not a small challenge. Traditionally, when a single machine learning model is used to predict air quality, it is difficult to obtain good prediction results under various AQI fluctuation trends. In order to solve this problem effectively, a prediction model based on Hierarchy Belief Rule Base-RFI is proposed in this paper, which provides a useful method for air quality prediction. First of all, in order to improve the accuracy of the model, random forest is used to screen the key features, and ER fusion is carried out on the features that are not screened. Secondly, the interpretability criteria of air quality prediction methods are proposed to standardize the interpretability of the whole modeling process. Then the inference and optimization methods of the model are given. Finally, the validity of the model is verified by experiments.
在交通标志检测任务中,交通标志的准确识别对于自动驾驶和智能交通系统至关重要。本文对Yolov8s模型改进并在CCTSDB2021数据集中进行实验以评估模型的性能,我们采取了以下改进,引入了专门针对小目标设计的检测头,该检测头通过优化特征图的尺度,增强了模型对小尺寸交通标志的识别能力。采用了SIoU损失函数,有助于提升小目标的定位精度。使用模块Star Block集成了一个轻量化模块LB,该模块通过减少参数量同时保持模型检测能力。在CCTSDB2021数据集上的实验结果表明,经过以上改进的Yolov8s模型在检测小尺寸交通标志时,不仅提高了检测准确率,而且减少了一定的参数量,展现了模型改进的有效性。In the traffic sign detection task, accurate recognition of traffic signs is crucial for autonomous driving and intelligent transportation systems. In this paper, we improve the Yolov8s model and conduct experiments in the CCTSDB2021 dataset to evaluate the performance of the model. We have made the following improvements: introducing a detection head specifically designed for small targets. This detection head enhances the model’s recognition ability for small-sized traffic signs by optimizing the scale of the feature map. The SIoU loss function is adopted, which helps to improve the positioning accuracy of small targets. Using the module Star Block, we integrated a lightweight module LB. This module reduces the number of parameters while maintaining the model’s detection ability. The experimental results on the CCTSDB2021 dataset show that the improved Yolov8s model not only improves the detection accuracy but also reduces a certain number of parameters when detecting small-sized traffic signs, demonstrating the effectiveness of model improvement.
准确预测处理器性能对计算机硬件设计与改进有着重要意义。然而,处理器预测系统存在两个核心问题:预测过程中处理器内部构造复杂和不确定性以及预测结果的不可解释性。置信规则库作为一种基于IF-THEN规则的建模方法,具有一定的可解释性并且可以处理复杂系统评估与预测中的不确定信息。但BRB的规则爆炸问题限制了专家知识的使用。因此,本文提出了一种基于近似置信规则库(ABRB)的处理器性能预测模型。该模型通过构建单属性BRB模型来解决规则爆炸问题,并通过基于投影协方差矩阵自适应进化策略(P-CMA-ES)算法对专家知识给定的初始参数进行优化。最后以UCI中处理器数据集为例,验证了所提方法的有效性。Accurate prediction of processor performance is important for computer hardware design and improvement. However, there are two core problems in processor prediction systems: the complexity and uncertainty of processor internals during the prediction process and the non-interpretability of the prediction results. Belief rule base (BRB), as a modelling method based on IF-THEN rules, has some interpretability and can handle uncertain information in the evaluation and prediction of complex systems. However, the rule explosion problem of BRB limits the use of expert knowledge. Therefore, this paper proposes a processor performance prediction model based on approximate belief rule base. The model solves the rule explosion problem by constructing a single-attribute BRB model and optimizes the initial parameters given by the expert knowledge by the Projection Covariance Matrix Adaptive Evolutionary Strategy (P-CMA-ES) based algorithm. Finally, the effectiveness of the proposed method is validated using the UCI mid-processor dataset as an example.
云制造系统(Cloud Manufacturing System, CMS)的网络安全对当今制造业影响重大,因此对于云制造系统进行网络安全态势预测(Network Security Situation Prediction, NSSP)变得十分重要。本文提出了一种基于证据推理(Evidential Reasoning, ER)和分层置信规则库(Hierarchical Belief Rule Base, HBRB)的云制造系统网络安全态势预测模型。首先,分析了影响CMS网络安全状况的因素,建立了评估框架,并采用ER算法进行融合,推导出CMS的安全态势值。其次,构建了基于HBRB的云制造系统网络安全态势预测模型,避免了属性过多引起的组合爆炸问题。此外,还使用了一种鲸鱼优化算法(Whale Optimization Algorithm, WOA),用于优化预测模型中的参数。该模型能够充分利用不确定信息与半定量信息,解决了专家知识的不完备性,提高了该预测模型的准确率。The cyber security of Cloud Manufacturing System (CMS) has a significant impact on today’s manufacturing industry, so it becomes important to perform Network Security Situation Prediction (NSSP) for CMS. In this paper, we propose a network security situation prediction model for CMS based on Evidential Reasoning (ER) and Hierarchical Belief Rule Base (HBRB). First, the factors affecting the cybersecurity status of CMS are analyzed, an assessment framework is established, and the ER algorithm is fused to derive the security posture value of CMS. Second, a cybersecurity posture prediction model for cloud manufacturing systems based on HBRB was constructed to avoid the combinatorial explosion problem caused by too many attributes. In addition, a Whale Optimization Algorithm (WOA) is used to optimize the parameters in the prediction model. The model can make full use of uncertain and semi-quantitative information, which solves the incompleteness of expert knowledge and improves the accuracy of this prediction model.
暂无评论