The exploration and resolution of persistent noise incursions within the tracking sequences, especially the occlusion, illumination variations, and fast motion, have garnered substantial attention for their functional properties in enhancing the accuracy and robustness of visual object trackers. However, existing visual object trackers, equipped with template updating mechanisms or calibration strategy, heavily rely on time-consuming historical data to achieve optimal tracking performance, impeding their real-time tracking capabilities. To address these challenges, this paper introduces a long-short term dual level memory augmented transformer structure aided visual object predictor (MeAP). The key contributions of MeAP can be summarized as follows: 1) the formulation of a noise model for specific invasion events based on incursion effects and corresponding template strategies serving as the foundation for more efficient memory utilization; 2) The memory exploration scheme based online tracking mask-based feature extraction strategy and the transformer architecture is introduced to mitigate the impact of noise invasion during memory vector construction; 3) the memory utilization scheme based target basic feature and dual feature target mask predictor is provided to implement the scene-edge feature for mask-based feature extraction method and jointly predict the accurate location of the tracking target.. Extensive experiments conducted on OTB100, NFS, VOT2021, and AVisT benchmarks demonstrate that MeAP, with its introduced modules, achieves comparable tracking performances against other state-of-the-art (SOTA) trackers, and operates at an average speed of 31 frames per second (FPS) across 4 benchmarks.

Most of the existing image fusion methods are dedicated to extracting the respective private features in the source image to reconstruct to get a fused image that contains rich information. However, its neglect of the enhancement of texture details in the source image as well as the preservation of background structures, which makes the fusion result look less impressive. To this end, an infrared and visible image fusion method based on a semi-global weighted least squares (SGWLS) method and guided edge-aware filter (GEAF) is proposed. First, a SGWLS method is applied to decompose the source image into base and detail layers containing different scale information. Then, the detail layer are texture-enhanced by using the scale coefficients. Due to the lack of edge features in the original base layer, for this reason, a guided edge-aware filter is designed, in which the enhanced detail layer was used as the guidance image as a way to perform a quadratic feature enhancement fusion for the base layer in order to retain richer gradient information. Finally, reconstruction of the fused detail and base layers by using inverse transformation to obtain the fusion result. Compared with nine state-of-the-art fusion algorithms, a large number of experimental results demonstrate the superior performance of the proposed method in highlighting texture details and background features, as well as the obvious advantages in SF and AG metrics.

Glaucoma is a prevalent eye disease that is responsible for causing blindness worldwide. To diagnose glaucoma, the cup-to-disc ratio (CDR) is an important factor. We introduce MRSNet, a novel segmentation network that incorporates encoding and decoding structures. The key innovation of this network is the application of convolutional block with large kernel convolutional attention to the task of medical image segmentation for optic cup and disc. By combining the benefits of residual and self-attention, our network achieves improved performance. The coding region of the network utilizes convolutional block with large kernel convolutional attention, enabling the extraction of multi-scale features with lower computational resources while also enhancing spatial attention. The self-attention layer acts as a transition between the encoding and decoding regions, capturing long connection information and providing additional image details. To further enhance segmentation performance, we employ a multi-resolution image combination approach and adaptively extract the input form using the compression and excitation module. Additionally, we propose a novel approach that combines the principle of consistency of deep supervision mechanisms with cross-entropy and Dice loss to guide the network towards accurate segmentation. In this study, we utilized a five-fold cross-validation method to train our network model. We then performed experimental validation and evaluation on three widely-used datasets, namely REFUGE, DRISHTI-GS, and RIM-ONE-r3. Our model achieved impressive results in the cup-to-disc ratio metric, which accurately reflects the segmentation effect. Specifically, we achieved scores of 0.0242, 0.0941, and 0.0158 for the aforementioned datasets, respectively. These scores outperformed some current classical algorithms. The experimental results demonstrate that the method proposed in this paper has the capability to extract more comprehensive information about the optic cup and disc, with the ability to generalize across different datasets. Furthermore, it shows that the convolutional block with large kernel convolutional attention module can be effectively utilized for the segmentation task of optic cup and disc. These findings provide a valuable research foundation for future researchers.

为了实现眼底图像视杯视盘的精准分割,减少人工分割方法带来的不确定性和耗时性,本文提出了 一种新型的卷积神经网络用于联合视杯视盘的分割,称为M2DS-TransUNet.该网络采用一种多分辨率图像结合并通过压缩与激励模块进行自适应提取的输入形式,同时结合多分辨率模块、Transformer和深度监督机制的优势,使得网络可以提取更加丰富的图像信息.采用五折交叉验证的方式对网络模型进行训练,并在当前三个主流数据集REFUGE、DRISHTI-GS和RIM-ONE-r3上进行了实验验证与评估,在最能体现分割效果的杯盘比指标上分别达到了 0.0284、0.097 8和0.017 9,其分割效果优于当前的一些经典算法.实验结果表明,本文所提出的方法可以提取更为丰富的视杯视盘信息,且具有跨数据集的泛化能力,是一种非常有竞争力的眼底图像视杯视盘联合分割方法.

To solve the problems of waste of human resources and cost in the development of logistics industry, an intelligent robot is designed to carry out material handling tasks with simple operation and high efficiency. Based on the multi-sensor three-degree-of-freedom manipulator, combined with STM32, OpenMV, motor and drive modules, PID closed-loop control, image processing, data analysis, power electronics, mechanical principle and other technologies are used to design and implement it. The results show that the device has the advantages of simple operation, stable performance, intelligent grasping and placing, etc. It can meet the needs of intelligent life and further improve people's quality of life.

Given the disconnection between the experimental content and the theory in the experimental project in the field of control engineering education, and the experimental process only stays at the level of virtual simulation and small development kit, we propose an intelligent manufacturing teaching assistant experimental platform based on the reconfigurable module. The “intelligence + reconfigurable modularization” experimental module is designed by simulating the industrial production line, and a high-fidelity experimental platform for control engineering education is assembled. We describe in detail the design and implementation process of the teaching assistant experimental platform and expound on the experimental project’s task design, technological revolution, and program design through a case study, which shows that the teaching auxiliary experimental platform is closely integrated with engineering practice. It is proved that students can acquire industrial technical ability from the experimental platform and improve their systematic cognitive ability in the manufacturing system.

文章以面向机电类专业的工程训练实训课程为教学改革对象,结合"赛课结合"思想和CDIO工程教育创新理念,探索了面向新工科创新型人才的实践类课程教学改革方法.从实训项目选取、实训模式重构、课程考核标准量化等方面进行了比较全面的改革与探索.通过对教学效果调查分析,所提出的教学改革方法取得了良好的效果.

以金属接线盒为研究对象,结合成形制件形状和大批量生产要求,分析并确定了合理的冲压工艺方案,采用正交试验法,利用Dynaform软件模拟分析了不同圆角半径对成形制件拉深性能的影响,确定并设计了落料拉深复合模与切边冲孔复合模,确保制件顺利成形.设计的模具结构合理,可为其他同类制件的成形提供参考.

Image inpainting is an essential issue in the field of computer vision. The basic purpose is to automatically recover the lost content according to the known content in the image. Although significant progress has been made in the completion of the regular missing image, the completion of the irregular image is still challenging. The completed images generated by the previous methods were different from the surrounding areas, considering the low efficiency of convolution in processing spatial position information. Based on this study, a partial convolution attention mechanism for image inpainting was proposed (PCNet), and an attention mechanism was introduced to extract long-distance and irregular image content which is a self-supervised module capable of focusing on global information. In addition, attentive normalization is introduced to model the long-distance relationship on the conditional image generation task. Experiments show that the results generated by our method are more natural and real, and the completed parts demonstrate more connected consistency.

新工科背景下,受现代行业运作与发展需求影响,高校工程教育应当重视学生的综合素质培育工作.但实际上很多高校依旧采用传统教学模式来培育学生,导致工科人才普遍存在专业素质突出、综合素质偏低的现象,这样一来不仅导致学生面临就业难度大,还造成行业运作与发展受限,高校工程教育模式需要改革与创新.该文对此将展开相关研究工作,阐述现代高校工程教育现状问题与成因,再围绕问题提出改革创新策略.

In the information era, the technology of biological character recognition has attracted more and more attentions. In this paper, by investigating theories of active appearance model and inverse compositional image alignment algorithm, we mainly proposed a semi-active appearance model for face alignment based on improving the classical models in the aspects of computation complexity, easily suffering from light, angle and expression, and so on. Firstly, the model of active appearance and the algorithm of alignment are investigated. For the inefficiency of classic gradient descent method in the matching process, the inverse compositional image alignment algorithm is proposed. Then, through combining the active appearance model and Grey Level Co-occurrence Matrix, a novel semi-active appearance model is proposed, which has a simple calculation and higher accuracy of face alignment. Finally, experiments were designed to demonstrate the effectiveness of the proposed algorithms.

In the context of regression to engineering practice,three typical problems of robot courses in higher engineering education in China are presented in this article. They are discipline-crossing,subject frontiers and engineering practice. In order to address these problems,this paper demonstrates how to construct hierarchical teaching system,diversify practice teaching links,present special lectures and step research teams combining with our practices in aspect of educational idea,teaching contents and practical effect. These all indicate our positive and beneficial exploration and attempt. Practice shows that undergraduate's engineering ability and innovation consciousness can be improved with integration of multi-faceted engineering perception,multi-crossed disciplines and diversified practical teaching process based on technology of intelligent robot,and the effect is more clear in the actual projects through mutual participation of teachers and students. Moreover,our efforts can also contribute on the valuable exploration and practice with respect to the cultivation of innovation-oriented engineering talents based on robotic education in universities and colleges.

Concerning the issue of how to achieve effective estimation of gravitational acceleration for attitude estimation of Micro Air Vehicle(MAV) under all dynamic conditions,an improved explicit complementary filter was proposed in combination with stepped-gain schedule.In order to validate the nonlinear complementary filter in the case when a MAV circled for an extended period,a centripetal acceleration mode was built using gyroscopes and indicated airspeed data,which resulted in precise estimation based on the estimation of gravitational acceleration and avoided reconstructing an estimation of the attitude.In the phase of Proportional-Integral(PI) compensation,the proportional gain and integral gain can achieve better adaptability by assigning different cut-off frequency value to the estimation of pitch and roll angles respectively.The experimental results show that the attitude angle estimation can be maintained under the range of ±2°.Compared with the typical filter algorithm,a better performance was achieved with respect to the efficiency and estimation error,so the algorithm proposed in this paper can be applied to accurate attitude estimation for MAV with low-cost inertial measurement units.

On the establishment of kinematics model of three-wheel omni-directional mobile robot,the basic forms and characteristics of linear and circular motion are analyzed and simulated,and distribution spaces of velocity and acceleration of translational linear motion of the robot are achieved.On the analysis of circular motion,the two forms of translation and rotation are discussed,furthermore,under the circumstance of circular motion with rotation,the conclusion is obtained and verified by simulation that the time optimal path between two points is a curved path but not a straight-line,which would be benefit on the improvement of robot's maneuverability and the realization of local optimal path planning in the competition of robot soccer.

In order to control the balance of the first-order inverted pendulum,which has nonlinear,coupled,multivariable and unstable system,the mathematic model is made through analysing the system,during the process of designing the Fuzzy PID controller,analyze the control theory of the Fuzzy PID controller;determine variables of fuzzy statements and sub-ordinative functions,stipulate the methods of fuzzy rules,and create solutions of the fuzziness.Finally make simulation studies on the first-order inverted pendulum by using Matlab Simulink and Fuzzy toolbox.Experimental results show that the Fuzzy-PID controller has very good control effect.

As information technology continues to progress, images, video and other multimedia files are widely transmitted over internet. Security threats like Viruses, Trojans, hackers hinder the application with data especially image exchange. Regarding this problem, an image sharing algorithm is proposed, and the matrix multiplication is used for image slicing. The original image can be recovered only when more than certain number of encrypted slices are collected and processed. The simulation experiment proves that the algorithm proposed can recover the image without loss of quality from qualified slices and any slices less than the threshold can not recover the image.

西南科技大学智能机器人创新实践班拔尖创新人才培养的实践表明,普通高校培养拔尖创新人才应结合本校办学定位和教育教学条件,遵循以生为本、因材施教的工作思路;创新教育理念,创造有利于人才成长的软硬件条件;引导学生树立正确的世界观、人生观和价值观,把知识传授、能力培养、素质提高贯穿于教育教学的全过程。

针对我国当前高等工程教育存在的问题,提出了以工程对象为纽带的工程实践模式,在寻求以激发学生兴趣为有效途径的基础上,构建了四层次的实践教学体系,探索出在"大工程"背景下培养具有较强业务理论基础、技术创新能力和深厚的人文社科素养的创新型工程人才新途径。

Based on the innovation and practice course system of this school,how to develop students＇ ability of innovation and practice has researched.According to structure of knowledge and expectation of students,the course＇s content has been established,and assessment and evaluation mode and teaching method has been discussed,and teaching process has also been analyzed.The implementation shows that in the course of practice,students are able to apply learned knowledge and teamwork to solve problems and increase ability of practice and innovation ability,and achieve the desired effect of the course.

Tackling of uncertain data is a major problem in data analysis and processing. The fuzzy theory with fuzzy numbers and fractal interpolation is employed to solve the issue of uncertainty. Sample data is used as the kernel of Gaussian fuzzy membership function and its fuzzy numbers are obtained by specifying λ-cut. These fuzzy numbers are used as uncertain data and defined as a new kind of fuzzy interpolation points. With these interpolation points fractal interpolation method is applied to fit curve of sample data. By these definitions, the flow of interpolation approach is given, and example is illustrated to show that a novel interpolation scheme is proposed for manipulating uncertain data.