Table of Contents

31 August 2021, Volume 42 Issue 4 Previous Issue    Next Issue
---
Contents
Review
Image Processing and Computer Vision
Computer Graphics and Virtual Reality
Digital Design and Manufacture
Industrial Design
Engineering Graphics
Total to Discuss


For Selected:

Download Citations EndNote Ris BibTeX

Toggle Thumbnails

Contents

Select

Table of Contents for Issue 4, 2021 2021, 42(4): 0-0.  Abstract ( 198 )   PDF (246KB) ( 133 )   CONTENTS Review Review of real-time deep learning-based object detection for mobile augmented reality   ·············································································································· GAO Wen-ting, LIU Yue (525) Survey on triangular mesh surface conformal parameterization   ························································· LI Hai-sheng, CAO Guo-liang, WEI Yang, WU Xiao-qun, CAI Qiang (535) Image Processing and Computer Vision Improved YOLOv4 small target detection algorithm with embedded scSE module   ······························································ JIANG Rong-qi, PENG Yue-ping, XIE Wen-xuan, XIE Guo-rong (546) Image super-resolution reconstruction algorithm based on dense residual network   ···················································································MAN Kai-liang, WANG You-sheng, LIU Ji-rong (556) Scale adaptive target tracking algorithm based on depth estimation and feature fusion   ··················································································· MU Qi, ZHANG Han, HE Zhi-qiang, LI Zhan-li (563) Classification algorithm of main bearing cap based on deep learning   ··························································· ZHANG Peng-fei, SHI Zhi-liang, LI Xiao-yao, OUYANG Xiang-bo (572) Multi-lead heartbeat signal classification based on multi-scale feature extraction   ···············································································ZHANG Fan, YIN Xin, XU Yu-yang, HAO Peng-yi (581) Multi-scale mural restoration method based on edge reconstruction   ········································································· ZHANG Hao-yuan, XU Dan, LUO Hai-ni, YANG Bing (590) Computer Graphics and Virtual Reality Interactive tree segmentation and modeling from ALS point clouds   ·········ZHANG Yue-kun, YU Wen-jie, ZHAO Xi-zhi, LYU Yan-xing, FENG Wei-huan, LI Zheng-rong, HU Shao-jun (599) A shape matching algorithm based on geometric moments and its applications on CAD models   ····································································· LENG Jue-lin, ZHANG Zhe, LIU Tian-tian, ZHENG Peng (608) Orthodontic path planning based on improved multi-PSO ···························· MA Tian, YANG Qin, LI Zhan-1i (615) Surface boundary shape curve fairing and simplification based on wavelet decomposition   ···························································································· HE Chuan, ZHAO Gang, WANG Ai-zeng (623) Real-time virtual and real occlusion processing technology based on voting decision   ································································· JIANG Su-qin, ZHANG Meng-jun, LI Wei-qing, SU Zhi-yong (629) Boundary pressure treatment and correction of surface particle position of solid entering water   ······························································· ZHU Xiao-lin, HE Hong-hong, GUO Qing-wei, ZHOU Yun-ruo (636) G1 interpolation algorithm and application of Bézier curves with monotone curvature   ·········································································· HE Chuan, ZHAO Gang, WANG Wei, WANG Ai-zeng (644) Geometric evaluation algorithms for Lupaş q-Bézier curve and its applications   ······························································································ LIU Li-hong, ZUO Hua, HAN Li-wen (651) Blended coordinates based on derived polygons ··························· HUANG Ru, LI Ya-juan, DENG Chong-yang (659) Digital Design and Manufacture Research on parametric design technology of high speed cam mechanism with oscillating follower   ··············································································· CAI Xiao-yao, CHE Wen-gang, CHEN Wen-bing (664) Industrial Design The optimal decision-making model of industrial design scheme based on multi-level grey comprehensive evaluation    method and its application·   ···LI Xue-rui, HOU Xing-gang, YANG Mei, HE Jia-xing, WANG Yi-yan, GUO Hao-yue, LI Xin-ying, WANG Lu-yao (670) Kansei evaluation method of product form design based on hesitant fuzzy linguistic term sets and particle    swarm optimization································································································ YANG Yan-pu (680) Evaluation method of vehicle side modeling based on neural network   ············································································ WANG Huan-huan, CHU Sheng-nan, GU Jing-wei (688) Engineering Graphics A new mixed teaching system based on OBE for cartography course   ·························································· WANG Yan, DU Xiu-hua, YANG Rui, ZHU Juan, CAO Xi-cheng (696) Related Articles | Metrics

Review

Select

Review of real-time deep learning-based object detection for mobile augmented reality GAO Wen-ting , LIU Yue 2021, 42(4): 525-534.  DOI: 10.11996/JG.j.2095-302X.2021040525 Abstract ( 473 )   PDF (2528KB) ( 490 )   Mobile augmented reality (AR) is a technology that integrates virtual information with the real world on the mobile intelligent terminal, therefore the ability to accurately detect the to-be-enhanced objects in the environment directly determines the performance of mobile AR systems. With the rapid advancement of deep learning, a large number of deep learning-based methods have been proposed for better detection. However, such problems as limited computing power, high energy consumption, large model size, and offloading latency make it difficult to combine deep learning-based object detection with mobile AR. This paper first summarized previous studies on deep learning-based object detection from both aspects of two stages and one stage, then categorized the object detection systems for mobile AR, and analyzed the approaches based on local, cloud, or edge ends, as well as collaboration. Finally, both the advantages and limitations of these methods were summarized, and predictions were made on the problems to be solved and the future development of object detection in mobile AR. Related Articles | Metrics

Select

Survey on triangular mesh surface conformal parameterization LI Hai-sheng , , CAO Guo-liang , , WEI Yang , , WU Xiao-qun , , CAI Qiang 2021, 42(4): 535-545.  DOI: 10.11996/JG.j.2095-302X.2021040535 Abstract ( 320 )   PDF (4512KB) ( 342 )   Surface parameterization is a significant topic in computer graphics and computer-aided design, with a wide range of applications. Surface parameterization aims to seek the minimization of some intrinsic geometry distortion (such as edge length, angle, and area) from 3D surface to a parameterization plane transformation. Angle-preserving parameterization preserves the local shape of the surface before and after parameterization, also known as conformal parameterization. This survey introduced the recent progress on triangular mesh surface conformal parameterization from computer graphics application angle. Based on different parameter domains, it mainly discussed planar parameterization, spherical parameterization, and hyperbolic parameterization. Finally, this survey presented a list of open research problems and possible future research directions. Related Articles | Metrics

Image Processing and Computer Vision

Select

Improved YOLOv4 small target detection algorithm with embedded scSE module JIANG Rong-qi , , PENG Yue-ping , XIE Wen-xuan , XIE Guo-rong 2021, 42(4): 546-555.  DOI: 10.11996/JG.j.2095-302X.2021040546 Abstract ( 435 )   PDF (1776KB) ( 377 )   To tackle the problems of low accuracy, high error rate, and high missed rate for small targets in target detection tasks, an improved YOLOv4 small target detection algorithm named scSE-IYOLOv4 was proposed. The experiment employed the VEDAI small target dataset and the K-means++ algorithm to optimize the anchor frame of the target sample, thereby improving the accuracy of the algorithm. Based on the YOLOv4 algorithm, studies and analyses were respectively conducted concerning the effect of the scSE attention module embedded in the model’s different positions and that of the addition of the SPP module to the model’s neck on the algorithm’s detection performance. The experiments proved that embedding the scSE attention module after the “Add” and “concat” layers of the backbone network of the YOLOv4 algorithm model, and adding the SPP module to the model’s neck can enhance the algorithm’s detection accuracy for small targets. The results showed that the mAP@0.5 both increased by 2.4% on the VEDAI test set. Finally, the scSE-IYOLOv4 algorithm was proposed based on the above experimental results of the improved backbone network and neck of the YOLOv4 model. It was proved by the experiment that scSE-IYOLOv4 can significantly heighten the detection accuracy for small targets. The value of mAP@0.5 on the VEDAI test set increased by 4.1%, compared with YOLOv4, and rose by 2.2% on the PASCAL VOC dataset. Related Articles | Metrics

Select

Image super-resolution reconstruction algorithm based on dense residual network MAN Kai-liang, WANG You-sheng, LIU Ji-rong 2021, 42(4): 556-562.  DOI: 10.11996/JG.j.2095-302X.2021040556 Abstract ( 216 )   PDF (2701KB) ( 368 )   Image super-resolution is a cheap and convenient method for image enhancement, which is of great significance in the research on video surveillance, medical imaging, satellite remote sensing, and other fields. In this paper, combined with the performance advantages of deep learning in image reconstruction, an enhanced residual dense network (ERDN) was proposed. Firstly, the enhanced residual dense block with multiple convolution kernels was employed to extract the details of the image. Then the multi-layer image information was filtered and reorganized through the skip connection and feature multiplexing modules, so that the network model could make comprehensive use of image information at different depths. Finally, the parameter vectors of the reconstructed model were constrained, and the decoupling operation of length and orientation enabled the model to converge at a larger learning rate to improve the model training speed. Experimental validation on multiple international public datasets shows that the method achieves better subjective visual evaluation and objective quantitative evaluation. For example, for the quadruple reconstruction task, the peak signal-to-noise ratio (PSNR) metric of the ERDN output image was improved by 0.24 dB over the dense residual network (RDN) on the Urban100 dataset, and the number of model parameters was reduced by about 50%, which can be applicable to super-resolution reconstruction of various scene images. Related Articles | Metrics

Select

Scale adaptive target tracking algorithm based on depth estimation and#br# feature fusion MU Qi , , ZHANG Han , HE Zhi-qiang , LI Zhan-li 2021, 42(4): 563-571.  DOI: 10.11996/JG.j.2095-302X.2021040563 Abstract ( 204 )   PDF (3922KB) ( 291 )   The correlation filter target tracking algorithm kernelized correlation filters (KCF) utilizes a single feature to describe the target, which cannot accurately track the target in complex environments, or when the target scale changes significantly. To address these problems, a scale adaptive target tracking algorithm based on depth estimation and feature fusion was proposed. Firstly, the depth of the target in the video was estimated using the depth estimation network, and the depth-scale estimation model was built and trained. In the tracking, the histogram of oriented gradient (HOG) features and color name (CN) features of the target were fused to train the correlation filter, and the depth of the target was obtained using the depth estimation network. In addition, the depth-scale estimation model was employed to obtain the scale of the target. Thus, the target size can be adjusted when the target scale changes, and the scale adaptive target tracking algorithm is realized. The experimental results show that compared with the classical algorithm KCF, higher accuracy can be obtained, and compared with the scale adaptive algorithm discriminative scale space tracking (DSST), the tracking speed is faster when the scale changes greatly, and the robustness is better in cases of complex environments and obscured targets. Related Articles | Metrics

Select

Classification algorithm of main bearing cap based on deep learning ZHANG Peng-fei , SHI Zhi-liang , LI Xiao-yao , OUYANG Xiang-bo 2021, 42(4): 572-580.  DOI: 10.11996/JG.j.2095-302X.2021040572 Abstract ( 139 )   PDF (2111KB) ( 203 )   The automatic classification and recognition algorithm of mechanical parts has broad application prospects in the fields of intelligent industry and automatic processing. In the automatic classification of automobile engine main bearing cap parts, there are difficult problems such as multi-surface distribution of features and light sensitivity. A multi-branch feature fusion convolutional neural network (MFF-CNN) was designed. The two sub-network branches of the MFF-CNN can extract the features of the two surfaces of the main bearing cap respectively, and form the final part classification feature after feature fusion. In terms of network structure design, the MFF-CNN was based on a densely-connected convolutional neural network design. By enhancing the feature reuse between network layers, the parameter amount of the model was effectively reduced, and the problems of overfitting and gradient disappearance of deep networks can be alleviated under the condition of small sample size. The experimental results show that the MFF-CNN can attain the recognition rate of 91.6% on the image data set of the main bearing cap collected in practice, and it displays good robustness in terms of the problem of uneven illumination of the parts’ images in actual production. Related Articles | Metrics

Select

Multi-lead heartbeat signal classification based on multi-scale feature extraction ZHANG Fan , YIN Xin , XU Yu-yang , HAO Peng-yi 2021, 42(4): 581-589.  DOI: 10.11996/JG.j.2095-302X.2021040581 Abstract ( 223 )   PDF (1340KB) ( 259 )   Electrocardiogram (ECG) is important for clinical diagnosis of cardiovascular disease. Extracting effective features is the key to automatic diagnosis system. Nowadays, most research works only use single-lead or two-lead signals, which makes the extracted features too incomprehensive to achieve excellent classification performance. In order to obtain more comprehensive features and better classification results, we proposed a multi-lead heartbeat signal classification method based on multi-scale feature extraction (MSNet). Firstly, the method received multi-lead stacked heartbeats signal matrix as input. Then, three feature vectors with different scale information were extracted by convolution layers. Finally, classification was designed to decide to which category the input heartbeat belonged based on fused features. Three open source databases including MIT-BIH Arrhythmia Database, MIT-BIH Supraventricular Arrhythmia Database, and St Petersburg INCART 12-Lead Arrhythmia Database were utilized in this research. Experiments on these databases with 5-fold cross-validation strategy show the excellent performance of the proposed method. For “normal-abnormal” classification, more than 99% of the accuracy, recall, precision, and F1 score can be achieved. For “multi-category” classification, 99.5% of the average accuracy, average recall, average precision, and average F1 score can be achieved. Compared with the state-of-the-art methods, the proposed method exhibits better performance. Related Articles | Metrics

Select

Multi-scale mural restoration method based on edge reconstruction ZHANG Hao-yuan, XU Dan, LUO Hai-ni, YANG Bing 2021, 42(4): 590-598.  DOI: 10.11996/JG.j.2095-302X.2021040590 Abstract ( 157 )   PDF (14502KB) ( 303 )   Ancient Chinese murals, with a history of thousands of years, inevitably have been damaged to varying degrees, so the relevant restoration research is of great historical and cultural value. In the process of the traditional manual restoration directly affecting the mural itself, improper operations will cause protective damages. Therefore, digital virtual restoration was adopted, and a multi-scale mural restoration method based on edge reconstruction was proposed. Due to the scarcity of existing murals, the experimental data was obtained by collecting and sorting out different regions of Chinese murals. The freely cropped murals were filtered to smooth the image details and retain their edges, which provided good initialization conditions for the image segmentation of damaged areas, then automatic calculations were conducted on the mask to be repaired after image clustering. The repair process was based on edge reconstruction, texture features of damaged murals were extracted in a multi-scale space, and the available known information of the image was fully excavated to fill in its missing content. The experimental results show that the method is not limited by the type and degree of damages to the murals, the entire reconstruction process is more versatile and effective, and it can recover images with complete structure and clear texture, and achieve better repair results. Related Articles | Metrics

Computer Graphics and Virtual Reality

Select

Interactive tree segmentation and modeling from ALS point clouds ZHANG Yue-kun , YU Wen-jie , ZHAO Xi-zhi , LYU Yan-xing , FENG Wei-huan , LI Zheng-rong , HU Shao-jun 2021, 42(4): 599-607.  DOI: 10.11996/JG.j.2095-302X.2021040599 Abstract ( 202 )   PDF (3375KB) ( 296 )   The airborne lidar scanning (ALS) system provides the possibility of acquiring large-scale tree point clouds only from a single scan, which helps to achieve the structural parameter extraction of trees and geometric reconstruction at the landscape level with higher accuracy. However, it remains a challenge to accurately segment and model trees from ALS point clouds, due to the diversity of tree species and the complex topology of trees. Although the traditional point cloud-based automatic tree segmentation and modeling algorithm are efficient, there remain such problems as great segmentation errors and lack of robustness for the modeling algorithm, making it difficult to meet users’ need for the precise annotation of tree segmentation and of modeling results in deep learning. In order to solve the problems of automatic segmentation and modeling caused by the sparse and incomplete ALS point clouds, an interactive hierarchical segmentation method was proposed based on height mapping. The proposed method can extract a single tree point cloud from sparse point clouds, and then utilize the improved space colonization algorithm (SCA) to model trees by interactively adjusting the parameters of constraint angle, kill distance, and influence radius. Experimental results show that the proposed interactive segmentation algorithm can avoid the false segmentation arising from the minimum spanning tree and Normalized-Cut algorithm. The proposed interactive modeling algorithm can robustly generate tree models from the sparse and incomplete ALS point clouds, as well as preserving the features of the original point cloud by selecting the appropriate combination of SCA parameters. Related Articles | Metrics

Select

A shape matching algorithm based on geometric moments and its applications on CAD models LENG Jue-lin , , ZHANG Zhe , , LIU Tian-tian , , ZHENG Peng 2021, 42(4): 608-614.  DOI: 10.11996/JG.j.2095-302X.2021040608 Abstract ( 225 )   PDF (2763KB) ( 155 )   A shape matching algorithm based on geometric moments was proposed to recognize CAD entities with similar shape features. A set of geometric moment invariants under similarity transformation were utilized to describe the shape features of geometric entities, and the similarity between geometric entities was evaluated according to those shape feature descriptors. In order to improve the accuracy and efficiency of geometric moment calculation, the triangular facet data of CAD model was preprocessed, and a recursive algorithm was adopted to accelerate computations of geometric moments. Finally, the proposed algorithm was applied to the pick of similar geometric entities of CAD models. Using graphical user interface of CAE software, CAD entities with similar shape features were picked up in real time, and the effectiveness of the proposed algorithm was validated. Related Articles | Metrics

Select

Orthodontic path planning based on improved multi-PSO MA Tian, YANG Qin, LI Zhan-i 2021, 42(4): 615-622.  DOI: 10.11996/JG.j.2095-302X.2021040615 Abstract ( 108 )   PDF (3455KB) ( 242 )   Orthodontic path planning is an integral part of a virtual orthodontic system. Aiming at the multi-objective and high-dimensional path planning problem, an improved multi-particle swarm optimization (PSO) path planning method was proposed. Firstly, the multi-PSO was utilized to model and design the tooth path planning to solve the problem of the overly high orthodontic dimension. Secondly, the difficulties in physiological reconstruction of different types of teeth were considered based on Beta curve and dental arch depth, and the inertia parameter w of different particle swarms was optimized to refine the orthodontic treatment process of the traditional orthodontic system. Finally, the upper and lower limits of position updating were modified to narrow the path search range, reduce the collision during orthodontic process, and optimize the orthodontic effect. The experimental results show that this method is 38% more efficient than the single particle swarm optimization model, which is more consistent with clinical orthodontic treatment, and is more likely to achieve the ideal orthodontic effect after treatment. Related Articles | Metrics

Select

Surface boundary shape curve fairing and simplification based on#br# wavelet decomposition HE Chuan , ZHAO Gang , WANG Ai-zeng 2021, 42(4): 623-628.  DOI: 10.11996/JG.j.2095-302X.2021040623 Abstract ( 72 )   PDF (1030KB) ( 99 )   This paper proposed an algorithm for boundary shape curve fairing and simplification. Compared with the previous methods, the selected boundary shape curve was faired with non-uniform B-spline wavelets by the new approach without superfluous control points, because T-mesh can be allowed in the result surface. Some examples were also presented in this paper to demonstrate the effectiveness of the new algorithm. Related Articles | Metrics

Select

Real-time virtual and real occlusion processing technology based on voting decision JIANG Su-qin , ZHANG Meng-jun , LI Wei-qing , SU Zhi-yong 2021, 42(4): 629-635.  DOI: 10.11996/JG.j.2095-302X.2021040629 Abstract ( 67 )   PDF (2758KB) ( 178 )   Aiming at the problems of poor real-time performance and low accuracy faced by the current virtual and real occlusion processing technology based on depth information, a real-time virtual and real occlusion processing algorithm was proposed based on local area depth estimation and patch similarity-based fusion of noise points. The algorithm took the real scene video sequence as the input. Firstly, the depth information of sparse key points was estimated using the local area depth estimation algorithm through sparse reconstruction, and the sparse depth was imposed on the target area to limit the depth propagation to the surrounding pixels, so as to quickly restore the relative depth map of the target area. Then, the noise point voting fusion algorithm employed depth comparison to determine the front and back position relationship of the virtual and real objects, and voted and merged the pixels in the area based on the patch similarity and voting decision method. Finally, the fusion effect with real occlusion relationship was output. The experimental results show that the proposed algorithm can not only improve the real-time performance of virtual and real occlusion, but also obtain a good fusion effect under different spatial relationships between real and virtual scenes. Related Articles | Metrics

Select

Boundary pressure treatment and correction of surface particle position of solid entering water  ZHU Xiao-lin, HE Hong-hong, GUO Qing-wei, ZHOU Yun-ruo 2021, 42(4): 636-643.  DOI: 10.11996/JG.j.2095-302X.2021040636 Abstract ( 60 )   PDF (2086KB) ( 82 )   When the boundary force method, virtual particle method, or coupling force method based on the traditional smooth particle hydrodynamics (SPH) method is used to treat solids entering water, the particle density at the fluid-solid interaction interface is discontinuous, the pressure is unstable, and some fluid particles penetrate the solid boundary or separate at this boundary. In addition, the fluid surface will be rougher under force after the surface is broken. Aiming at the above problems, this paper combined the advantages of the boundary force method and virtual particle method, and improved the coupling force method to deal with the moving solid boundary to prevent fluid particles from penetrating the solid boundary. The pressure calculation method of the interactive interface was enhanced, the calculation accuracy was increased, and the pressure field of the interactive interface was stabilized. Corrections were made to the particle position on the fluid surface to improve the simulation effect of the free-flowing liquid surface boundary on the fluid surface. The proposed method was verified by the classic two-dimensional solid entering water experiments. The experimental results show that after fluid particles interact with solid particles, stable interface pressure can be maintained using the method in this paper, the interface separation is clear without penetration, the surface fluid particles are uniformly distributed, and the movement of the flow field is real and natural. Related Articles | Metrics

Select

G1 interpolation algorithm and application of Bézier curves with monotone curvature HE Chuan , ZHAO Gang , WANG Wei , WANG Ai-zeng 2021, 42(4): 644-650.  DOI: 10.11996/JG.j.2095-302X.2021040644 Abstract ( 93 )   PDF (670KB) ( 91 )   Based on the Bézier curve with monotone curvature, this paper presented an accurate and efficient spline interpolation algorithm satisfying G1 constraint. With the given two point-orientation pairs, the construction was to solve the nonlinear equations using the geometric design criteria of monotone curvature, so as to obtain the Bézier curve satisfying the G1 interpolation condition. Compared with the interpolation algorithm based on Euler spiral, the advantages of the proposed method are as follows: simple construction, accurate calculation, and compatibility with the existing NURBS. Based on the idea of piecewise, this algorithm can deal with the interpolation problems of a given sequence of points and the direction of head and tail tangents, manifesting stronger adaptability and universality. Related Articles | Metrics

Select

Geometric evaluation algorithms for Lupaş q-Bézier curve and its applications LIU Li-hong , ZUO Hua , HAN Li-wen 2021, 42(4): 651-658.  DOI: 10.11996/JG.j.2095-302X.2021040651 Abstract ( 53 )   PDF (448KB) ( 70 )   The Lupaş q-Bézier curve is a generalized Bézier curve with q-integer as the shape parameter. A new geometric evaluation algorithm for Lupaş q-Bézier curve was constructed in this paper, in which the affine combination of two nodes in the penultimate layer of the algorithm is tangent to the curve. A new representation of the derivative of the Lupaş q-Bézier curve was obtained using the property of algorithm, and the subdivision of curve was realized. Particularly, a product of the shape parameters of the two subcurves produced by the subdivision of the Lupaş q-Béziercurves was equal to the shape parameter of the original curve. Furthermore, a new geometric evaluation algorithm of weighted Lupaş q-Bézier with an explicit matrix representation was gained. Related Articles | Metrics

Select

Blended coordinates based on derived polygons HUANG Ru, LI Ya-juan, DENG Chong-yang 2021, 42(4): 659-663.  DOI: 10.11996/JG.j.2095-302X.2021040659 Abstract ( 46 )   PDF (1740KB) ( 204 )   Blended barycentric coordinates based on derived polygons were proposed based on the existing barycentric coordinates by triangulating the initial polygon, and deriving the point polygons, the edge polygons, and the face polygons of the triangle using a triangular mesh. First, the derived polygons were obtained using a triangular mesh. Then, the derived polygons’ vertices, which are about the initial polygons’ vertices, could be calculated by referring to harmonic coordinates, local barycentric coordinates, or iterative coordinates. Next, a point as an affine combination of the derived polygons’ vertices could be represented using iterative coordinates. The most important step was to represent the point as an affine combination of the initial polygon’s vertices using a suited mixing coefficient. Taking all the instances into consideration, the conclusion can be naturally drawn that the proposed construction of barycentric coordinates are local, non-negative, and smooth. Related Articles | Metrics

Digital Design and Manufacture

Select

Research on parametric design technology of high speed cam mechanism with oscillating follower CAI Xiao-yao , CHE Wen-gang , CHEN Wen-bing 2021, 42(4): 664-669.  DOI: 10.11996/JG.j.2095-302X.2021040664 Abstract ( 96 )   PDF (757KB) ( 91 )   Aiming at the problems of cam angle deviation caused by follower swing in the parametric design process of high-speed cam mechanism with oscillating follower, this paper analyzed the causes and effects of cam angle deviation, proposed a mathematical model of parametric design based on angle compensation, and realized the accurate mapping from cam parameters to cam profile curve. Taking the improved sinusoidal motion law as an example, parametric design and characteristic curve visualization of cam mechanism with oscillating follower were realized based on UG software and secondary development platform, and the accuracy of mathematical model and characteristic curve was verified by UG motion simulation. The results show that the center angle of the cam profile curve is different from the cam angle in push and return stages, and that the mathematical model based on the angle compensation can reduce the maximum acceleration of the cam follower, realize the accurate design of the cam profile curve, simplify the design and modification process of the cam swing mechanism, and greatly improve the design efficiency. Related Articles | Metrics

Industrial Design

Select

The optimal decision-making model of industrial design scheme based on multi-level grey comprehensive evaluation method and its application LI Xue-rui, HOU Xing-gang, YANG Mei, HE Jia-xing, WANG Yi-yan, GUO Hao-yue, LI Xin-ying, WANG Lu-yao 2021, 42(4): 670-679.  DOI: 10.11996/JG.j.2095-302X.2021040670 Abstract ( 211 )   PDF (3189KB) ( 122 )   In order to assist enterprises in making reasonable decisions during the evaluation stage of product design schemes and reducing blind subjectivity of design, a decision-making model for selecting industrial product design schemes was constructed based on the multi-level grey comprehensive evaluation method. In the evaluation process, firstly, the weight coefficient of the evaluation index was reasonably determined through qualitative and quantitative analysis according to the analytic hierarchy process (AHP) and entropy method. Secondly, the gray correlation analysis method was adopted to calculate the gray weighted correlation degree of each design plan, the order of the design schemes was determined, and the comprehensive evaluation and selection of the design schemes were completed based on the results. Based on this model, the priority ranking of five design schemes of a certain multi-rotor UAV was completed, and the results were verified according to the entropy weight TOPSIS method, proving the feasibility of this model. Finally, based on the index weights and program scores, the optimization of the selected scheme was conducted, and the scheme was visualized by three-dimensional modeling. Thus, the role in design assistance of the decision-making model for evaluating and selecting UAV products, based on the multi-level gray comprehensive evaluation method, was effectively demonstrated. Related Articles | Metrics

Select

Kansei evaluation method of product form design based on hesitant fuzzy linguistic term sets and particle swarm optimization YANG Yan-pu 2021, 42(4): 680-687.  DOI: 10.11996/JG.j.2095-302X.2021040680 Abstract ( 84 )   PDF (703KB) ( 79 )   The Kansei evaluation of product form design reflects users’ perceptual cognition, of which the characteristics of fuzziness and uncertainty render users hesitant and unable to precisely express their Kansei preference. To address this problem, the hesitant fuzzy linguistic term sets (HFLTSs) were introduced to describe users’ Kansei evaluation. With the mathematical operators of HFLTSs, a consensus model was constructed to measure the consistency degree of users’ perceptual cognition about product form design schemes. The particle swarm optimization technique (PSO) was utilized to optimize users’ Kansei evaluation matrices for consensus reaching. With optimized Kansei evaluation matrices, the technique for order preference by similarity to an ideal solution (TOPSIS) was applied to rank product form design schemes. A Kansei evaluation process of product form design based on HFLTSs and PSO was proposed. The Kansei evaluation of car charging pile design schemes as an example verified that HFLTSs can be conducive to solving the uncertainty and hesitancy problems of users’ perceptual cognition, and that by integrating PSO for consensus reaching, the consistency of Kansei evaluation with HFLTSs can be improved, and finally the quality of Kansei evaluation of product forms can be enhanced. Related Articles | Metrics

Select

Evaluation method of vehicle side modeling based on neural network WANG Huan-huan , CHU Sheng-nan , GU Jing-wei 2021, 42(4): 688-695.  DOI: 10.11996/JG.j.2095-302X.2021040688 Abstract ( 84 )   PDF (1336KB) ( 64 )   By using the back propagation (BP) neural network, the relationship between car styling contours and image semantics was constructed in MATLAB to quickly judge the vehicle side modeling style. Then the expression recognition model built by the convolutional neural network (CNN) was employed to establish the automobile model evaluation system, and to analyze and identify users’ preferences for the new design, thus obtaining the vehicle side modeling scheme which can meet users’ emotional needs. Finally, the feasibility of the method was verified through examples, and the optimal curvature range of the flow-type car was inferred. The experimental results show that the quantitative evaluation method of automobile modeling based on neural networks can evaluate the product modeling design more accurately and produce the side shape of concrete image in the form of data. Related Articles | Metrics

Engineering Graphics

Select

A new mixed teaching system based on OBE for cartography course WANG Yan, DU Xiu-hua, YANG Rui, ZHU Juan, CAO Xi-cheng 2021, 42(4): 696-702.  DOI: 10.11996/JG.j.2095-302X.2021040696 Abstract ( 207 )   PDF (1577KB) ( 152 )   Under the sudden strike of the pandemic, the online teaching of “no class, still learning” rendered the teaching under outcome-based education (OBE) special. Taking the course “Modern Engineering Graphics” as the carrier, online teaching that trained students’ graphic literacy and integrated the computer modeling was put forward. The teaching focused on integrating teaching resources, relying on the network platform, and cultivating the engineering ability. The teaching process emphasized the student-centered teaching philosophy and strengthened diversified appraisals. A new mixed system of cartography teaching was constructed, in-class and outside-class, online and offline, theoretical and practical. It was practiced in the majors of mechanical design and automation. The following problems were solved, such as poor interaction, difficulties in practice and cartography teaching, and the lack of actual computer operation in distance learning. Thus, the teaching effect equivalent to the offline course was attained. It can provide the reformative idea for the teaching of basic engineering courses. Related Articles | Metrics

Total to Discuss

Select

Vol.42, No.4, 2021 2021, 42(4): 703-703.  Abstract ( 30 )   PDF (53866KB) ( 49 )   Related Articles | Metrics