مشتقات ثابت دو بعدی تفکیک پذیر صریح برای تشخیص جسم
سال انتشار: 2019 - تعداد صفحات فایل pdf انگلیسی: 9 - تعداد صفحات فایل doc فارسی: 19
مشتقات ثابت تصویر به طور گسترده ای در زمینه های تشخیص الگو و دید رایانه مورد استفاده قرار گرفته اند، زیرا آنها قادر به ارائه الگوی ویژگی های مستقل تبدیل هندسی هستند. در حال حاضر، ثابت های تفکیک پذیر و مشتقات آنها به دلیل توانایی در ترکیب ویژگی های اساسی ثابت های متعامد مختلف، بیشتر مورد توجه قرار گرفته است. با این حال، بسیاری از مشتق های ثابت تفکیک پذیر موجود، به طور غیرمستقیم از مشتق های هندسی و بر اساس رابطه چندجمله ای متعامد و هندسی، به دست می آیند. بنابراین، در این مقاله، رویکرد مستقیمی برای ساخت مجموعه ای از مشتق های ثابت تفکیک پذیر گسسته Chebichef-Krawtchouk پیشنهاد شد که در آن به طور همزمان مشتق برای چرخش، مقیاس پذیری و تبدیل انتقال فراهم می شود و مبتنی بر فرم صریح چند جمله ای Tchebichef و Krawtchouk است. در نتیجه، نتایج تجربی و نظری اثربخشی روش پیشنهادی اثبات شد و ارجحیت آنها در طبقه بندی تصویر و شناخت الگو در مقایسه با روش های موجود نشان داده شد.
کليدواژه: مشتقات غیرمستقیم | روش صریح | ثابت تفکیک پذیر | چندجمله ای Krawtchouk | چندجمله ای Tchebichef | تشخیص الگو
|مقاله ترجمه شده|
Dynamic texture analysis with diffusion in networks
تجزیه و تحلیل بافت پویا با انتشار در شبکه-2019
Dynamic texture is a field of research that has gained considerable interest from computer vision community due to the explosive growth of multimedia databases. In addition, dynamic texture is present in a wide range of videos, which makes it very important in expert systems based on videos such as medical systems, traffic monitoring systems, forest fire detection system, among others. In this paper, a new method for dynamic texture characterization based on diffusion in directed networks is proposed. The dynamic texture is modeled as a directed network. The method consists in the analysis of the dynamic of this network after a series of graph cut transformations based on the edge weights. For each network transformation, the activity for each vertex is estimated. The activity is the relative frequency that one vertex is visited by random walks in balance. Then, texture descriptor is constructed by concatenating the activity histograms. The main contributions of this paper are the use of directed network modeling and diffusion in network to dynamic texture characterization. These tend to provide better performance in dynamic textures classification. Experiments with rotation and interference of the motion pattern were conducted in order to demonstrate the robustness of the method. The proposed approach is compared to other dynamic texture methods on two very well known dynamictexture database and on traffic condition classification, and outperform in most of the cases.
Keywords: Dynamic texture | Complex networks | Diffusion | Random walks
Weld image deep learning-based on-line defects detection using convolutional neural networks for Al alloy in robotic arc welding
تشخیص نقص تصویر جوش تشخیص عمیق مبتنی بر یادگیری عمیق برخط با استفاده از شبکه های عصبی همگرا برای آلیاژ آل در جوش قوس رباتیک-2019
Accurate on-line weld defects detection is still challenging for robotic welding manufacturing due to the complexity of weld defects. This paper studied deep learning–based on-line defects detection for aluminum alloy in robotic arc welding using Convolutional Neural Networks (CNN) and weld images. Firstly, an image acquisition system was developed to simultaneously collect weld images, which can provide more information of the real-time weld images from different angles including top front, top back and back seam. Then, a new CNN classification model with 11 layers based on weld image was designed to identify weld penetration defects. In order to improve the robustness and generalization ability of the CNN model, weld images from different welding current and feeding speed were captured for the CNN model. Based on the actual industry challenges such as the instability of welding arc, the complexity of the welding environment and the random changing of plate gap condition, two kinds of data augmentation including noise adding and image rotation were used to boost the CNN dataset while parameters optimization was carried out. Finally, non-zero pixel method was proposed to quantitatively evaluate and visualize the deep learning features. Furthermore, their physical meaning were clearly explained. Instead of decreasing the interference from arc light as in traditional way, the CNN model has taken full use of those arc lights by combining them in a various way to form the complementary features. Test results shows that the CNN model has better performance than our previous work with the mean classification accuracy of 99.38%. This paper can provide some guidance for on-line detection of manufacturing quality in metal additive manufacturing (AM) and laser welding.
Keywords: Deep learning | Defects detection | Al alloy | Robotic arc welding | Convolutional neural networks | Weld images | Feature visualization
An efficient colour image encryption scheme based on 1-D chaotic maps
یک طرح رمزگذاری کارآمد تصویر بر اساس نگاشت های هرج و مرج 1 بعدی-2019
Nowadays, internet is the medium through which people are sharing digital images. However, security is an issue in sharing of those images. Encryption techniques play an important role for the security of images. In this paper, an efficient colour image encryption scheme is proposed using multiple piece-wise linear chaotic map (PWLCM) systems. The proposed scheme first performs two-times multi-way block- based rotational permutation operations and then row-column rotational permutations. Finally, it per- forms row, column, and block diffusion operations. This scheme is simple and secure because it involves only simple rotational permutation operations using only PWLCM systems. Also, this scheme is efficient because it requires less computations and less simulation time to encrypt colour images. The advantages of this scheme are high key space, high security, simplicity and high efficiency in terms of computational complexity and simulation time. The simulation outputs and the security analyses indicate that the pro- posed method has better performance in terms of encryption efficiency, security, and resistivity for most of the common attacks.
Keywords: Security | Chaotic cryptography | Colour image encryption | Rotational permutation operation | PWLCM system | Efficiency | Key space
Equivalence of 2-rotation symmetric quartic Boolean functions
هم ارزی توابع بولی کوارتزی متقارن 2-چرخش-2019
A Boolean function in n variables is 2-rotation symmetric if it is invariant under even pow- ers of ρ(x 1 , . . . . . . , x n ) = (x 2 , . . . , x n , x 1 ) , but not under the first power (ordinary rotation symmetry); we call such a function a 2-function. A 2-function is called monomial rota- tion symmetric (MRS) if it is generated by applying powers of ρ2 to a single monomial. If the quartic MRS 2-function in 2 n variables has a monomial x 1 x q x r x s , then we use the notation 2-(1, q, r, s ) 2 n for the function. This paper gives a detailed theory of equivalence of quartic MRS 2-functions in 2 n variables. Such a theory was provided for the cubic MRS 2-functions in two 2015 papers of Cusick and Johns. As in the earlier papers, the two main topics in the theory are describing the affine equivalence classes of the functions under certain groups of permutations; and giving details of the linear recursions that the Ham- ming weights of any sequence of functions 2-(1, q, r, s ) 2 n (with q < r < s , say), n = s, s + 1 , . . . can be shown to satisfy. The discussion for both of these topics uses new ideas because the quartic theory naturally divides into two cases.
Keywords: Boolean | function Quartic | Rotation symmetric | Affine equivalence | Hamming weight | Cryptography
Fuzzy control system for variable rate irrigation using remote sensing
سیستم کنترل فازی برای آبیاری با سرعت متغیر با استفاده از سنجش از دور-2019
Variable rate irrigation (VRI) is the capacity to spatially vary the depth of water application in a field to handle different types of soils, crops, and other conditions. Precise management zones must be devel- oped to efficiently apply variable rate technologies. However, there is no universal method to determine management zones. Using speed control maps for the central pivot is one option. Thus, this study aims to develop an intelligent fuzzy inference system based on precision irrigation knowledge, i.e., a system that can create prescriptive maps to control the rotation speed of the central pivot. Satellite images are used in this study because remote sensing offers quick measurements and easy access to information on crops for large irrigation areas. Based on the VRI-prescribed map created using the intelligent decision- making system, the pivot can increase or decrease its speed, reaching the desired depth of application in a certain irrigation zone. Therefore, considering the spatial variability in the crop has made the strategy of speed control more realistic than traditional methods for crop management. The intelligent irrigation system pointed out areas with lower leaf development, indicating that the pivot must reduce its speed, thus increasing the water layer applied to that area. The existence of well-divided zones could be ob- served; each zone provides a specific value for the speed that the pivot must develop for decreasing or increasing the application of the water layer to the crop area. Three quarters of the total crop area had spatial variations during water application. The set point built by the developed system pointed out zones with a decreased speed in the order of 50%. From the viewpoint of a traditional control, the relay from pivot percent timer should have been adjusted from 70% to 35% whenever the central pivot passed over that specific area. The proposed system obtained values of 37% and 47% to adjust the pivot percent timer. Therefore, it is possible to affirm that traditional control models used for central-pivot irrigators do not support the necessary precision to meet the demands of speed control determined by the developed VRI systems. Results indicate that data from the edaphoclimatic variables when well-fitted to the fuzzy logic can solve uncertainties and non-linearities of an irrigation system and establish a control model for high-precision irrigation.
Keywords: Fuzzy control | Variable rate irrigation | Speed control | Remote sensing | Decision support system
Could LSA become a “Bifactor”model? Towards a model with general and group factors
آیا LSA می تواند به یک مدل Bifactor تبدیل شود؟ به سمت یک مدل با عوامل کلی و گروهی-2019
One insufficiently grounded criticism made against Latent Semantic Analysis is that it is impossible to semantically interpret its dimensions. This is not true, as several studies have transformed the latent se- mantic space to interpret them, by means of some methods. One of them is the Inbuilt-Rubric method. Rather than grouping concepts around dimensions, as in Exploratory Factor Analysis based rotation meth- ods, the Inbuilt-Rubric is a method that perform an “a priori”imposition of concepts onto the latent se- mantic space. It uses a confirmatory strategy. This study seeks to propose solutions for two limitations found in the current Inbuilt-Rubric methodology: one solution is inspired by Bifactor Models and the management of common variance of the concepts involved; and the other one is based in randomizing the sequence to perform the process. Both methods outperform the current Inbuilt-Rubric version in rel- evant content detection. The reported improvements can be incorporated into expert systems that use Latent Semantic Analysis and Inbuilt-Rubric in relevant content detection or text classification tasks.
Keywords: Latent semantic analysis | Bifactor model | Distributional semantics | Inbuilt-Rubric method | Rotation | Text assessment
Quasi-autonomous bolt-loosening detection method using vision-based deep learning and image processing
روش تشخیص شل شدن شبه مستقل با استفاده از یادگیری عمیق مبتنی بر بینایی و پردازش تصویر-2019
In this study, a quasi-autonomous vision-based method is newly proposed for detecting loosened bolts in critical connections. The main idea of the approach is to estimate the rotational angles of bolts from the connection images by integrating deep learning technology with image processing techniques. Firstly, a regional convolutional neural network (RCNN)-based deep learning algorithm is developed to automatically detect and crop plausible bolts in the connection image. Also, the Hough line transform (HLT)-based image processing algorithm is designed to automatically estimate the bolt angles from the cropped bolt images. Secondly, the proposed vision-based approach is validated for bolt-loosening detection in a lab-scale girder connection using images captured by a smartphone camera. The accuracy of the RCNN-based bolt detector and the HLT-based bolt angle estimator are examined under different levels of perspective distortion and shooting distance. Finally, the practicality of the proposed vision-based method is verified on a real-scale girder bridge connection containing numerous bolts. The images of the connection are captured by an unmanned aerial vehicle and transferred to a computer where a quasi-autonomous bolt-loosening detection process is performed via the proposed algorithm. The experimental results demonstrate potentials of the proposed approach for quasi real-time bolt-loosening monitoring of large bolted connections. The results show that the perspective angle should not go beyond 40 degrees to ensure the accuracy of the detection results.
Keywords: Bolted connection | Bolt-loosening | Deep learning | CNN | Hough transform | Canny line detector | Bolt detection | Bolt rotation estimation
Design of optimized soft soles for humanoid robots
طراحی در جهت بهینه شدن کف پای نرم در ربات های انسان نما-2017
Article history:Available online 31 May 2017Keywords:Shape optimizations Compliant soles Humanoid robots WalkingWe describe a methodology to design foot soles for a humanoid robot given walking gait parameters (i.e. given center-of-mass and zero-moment-point trajectories). In order to obtain an optimized compliant sole, we devised a shape optimization framework which takes — among other inputs, an initial rough (simplified) shape of the sole and refines it through successive optimization steps under additional constraints and a cost function. The shape is optimized based on the simulation of the sole deformation during an entire walking step, taking time dependent input of the walking pattern generator into account. Our shape optimization framework is able to minimize the impact of the foot with the ground during the heel-strike phase and to limit foot rotation in case of perturbations. Indeed, low foot rotation enforces a vertical posture and secures the balance of the humanoid robot. Moreover, weight restriction (formulated as a constraint on the sole volume) is added to our optimization problem.© 2017 Elsevier B.V. All rights reserved.1.
Keywords:Shape optimizations | Compliant soles | Humanoid robots | Walking
Exoskeletal master device for dual arm robot teaching
دستگاه اصلی Exoskeletal برای آموزش ربات دو محور-2017
Article history:Received 18 May 2016Revised 8 February 2017Accepted 26 February 2017Available online 17 March 2017Keywords: Master device Motion teaching Dual arm robot Exoskeleton Prismatic jointDual arm robots are used as humanoid or industrial robots for assembly work. Each arm of these robots is generally composed of 7-DOF to mimic the human arm. For motion teaching of this 7-DOF robot arm, upper limb exoskeletal master devices can be used, and each arm of the upper limb exoskeletal master device can also be composed of 7-DOF. However, the motions of the human shoulder are complex and the lack of DOF in the exoskeletal master device limits the wearer’s motions and makes the wearer feel uncomfortable. We propose a compact-sized exoskeletal master device, each of whose arms are com- posed of two serially connected parts. One is a 6-DOF shoulder–elbow mechanism and the other is a 3-DOF wrist mechanism. The 6-DOF mechanism serially connects the base frame near the shoulder to a point on the forearm, and the center of rotation of the shoulder joint is shifted to the outside of the human shoulder. In addition, the 6-DOF mechanism includes a long stroke but short reduced-length pris- matic joint. This 6-DOF mechanism enables unconstrained and comfortable shoulder motions. The 3-DOF wrist mechanism corresponds to forearm pronation/supination, wrist ﬂexion/extension, and wrist adduc- tion/abduction. The closed-loop inverse kinematic scheme is applied for the dual arm robot control in the task space. The performance of the exoskeletal master devices and control strategies are veriﬁed through experiments using a 14-DOF dual arm slave robot.© 2017 Elsevier Ltd. All rights reserved.1.
Keywords: Master device | Motion teaching | Dual arm robot | Exoskeleton | Prismatic joint