Computer-vision classification of corn seed varieties using deep convolutional neural network
طبقه بندی بینایی ماشین انواع بذر ذرت با استفاده از شبکه عصبی پیچیده عمیق-2021
Automated classiﬁcation of seed varieties is of paramount importance for seed producers to maintain the purity of a variety and crop yield. Traditional approaches based on computer vision and simple feature extraction could not guarantee high accuracy classiﬁcation. This paper presents a new approach using a deep convolutional neural network (CNN) as a generic feature extractor. The extracted features were classiﬁed with artiﬁcial neural network (ANN), cubic support vector machine (SVM), quadratic SVM, weighted k-nearest-neighbor (kNN), boosted tree, bagged tree, and linear discriminant analysis (LDA). Models trained with CNN-extracted features demonstrated better classiﬁcation accuracy of corn seed varieties than models based on only simple features. The CNN-ANN classiﬁer showed the best performance, classifying 2250 test instances in 26.8 s with classiﬁcation accuracy 98.1%, precision 98.2%, recall 98.1%, and F1-score 98.1%. This study demonstrates that the CNN-ANN classiﬁer is an efﬁcient tool for the intelligent classiﬁcation of different corn seed varieties.© 2021 Elsevier Ltd. All rights reserved.
Keywords: Machine vision | Deep learning | Feature extraction | Non-handcrafted features | Texture descriptors
Optimization of volleyball motion estimation algorithm based on machine vision and wearable devices
بهینه سازی الگوریتم برآورد حرکت والیبال بر اساس بینایی ماشین و دستگاه های پوشیدنی-2021
Volleyball is a team sport of track video-based analysis essential. However, this is difficult, especially in machine vision and wearable devices. Due to the ball’s small size, its speed of motion blur is generated, generally blocked. Older systems find it difficult to analyze the level of volleyball. When tracking with machine vision and wearable devices, it is recommended that Volleyball Motion Estimation Algorithms improve robustness to player confusion between different particle series. Faster processing speed and less confusing players presented herein as a method superior to the conventional particulate filter. The ball and Volleyball Motion Estimation Algorithm uses ma- chine vision and wearable devices to detect differential picture, and motion machine vision wearable device is optimized. Optimized toss Volleyball is correct by the Volleyball Motion Estimation Algorithm, and the trajectorys almost equal to the false prediction of the ball’s size. Machine vision and wearable devices can provide a new sports attendance watching experience through predictive images superimposed on live broadcasts. This also shows that the method can identify some important parts of the body to help predict thrown. Compared to the previous system, which provides better results.
Keywords: Volleyball motion estimation algorithm | Machine vision and wearable devices | Optimization of ball tracking | Classical particle filter | Predicted images
The quiet revolution in machine vision - a state-of-the-art survey paper, including historical review, perspectives, and future directions
انقلاب آرام در بینایی ماشین-مقاله ای پیشرفته مروری، شامل مرور تاریخی ، چشم اندازها و جهت های آینده-2021
Over the past few years, what might not unreasonably be described as a true revolution has taken place in the ﬁeld of machine vision, radically altering the way many things had previously been done and offering new and exciting opportunities for those able to quickly embrace and master the new techniques. Rapid developments in machine learning, largely enabled by faster GPU-equipped computing hardware, has facilitated an explosion of machine vision applications into hitherto extremely challenging or, in many cases, previously impossible to automate industrial tasks. Together with developments towards an internet of things and the availability of big data, these form key components of what many consider to be the fourth industrial revolution. This transformation has dramatically improved the efﬁcacy of some existing machine vision activities, such as in manufacturing (e.g. inspection for quality control and quality assurance), security (e.g. facial biometrics) and in medicine (e.g. detecting cancers), while in other cases has opened up completely new areas of use, such as in agriculture and construction (as well as in the existing domains of manufacturing and medicine). Here we will explore the history and nature of this change, what underlies it, what enables it, and the impact it has had - the latter by reviewing several recent indicative applications described in the research literature. We will also consider the continuing role that traditional or classical machine vision might still play. Finally, the key future challenges and developing opportunities in machine vision will also be discussed.© 2021 Elsevier B.V. All rights reserved.
Keywords: Machine vision | Machine learning | Deep learning | State-of-the-art
Computer vision based food grain classification: A comprehensive survey
طبقه بندی دانه های غذایی مبتنی بر بینایی رایانه ای: یک مرور جامع-2021
This manuscript presents a comprehensive survey on recent computer vision based food grain classification techniques. It includes state-of-the-art approaches intended for different grain varieties. The approaches pro- posed in the literature are analyzed according to the processing stages considered in the classification pipeline, making it easier to identify common techniques and comparisons. Additionally, the type of images considered by each approach (i.e., images from the: visible, infrared, multispectral, hyperspectral bands) together with the strategy used to generate ground truth data (i.e., real and synthetic images) are reviewed. Finally, conclusions highlighting future needs and challenges are presented.
Keywords: Computer vision approaches | Quality inspection | Food grain identification | Machine vision
A measurement method of motion parameters in aircraft ground tests using computer vision
روش اندازه گیری پارامترهای حرکت در آزمایش های زمینی هواپیماها با استفاده از بینایی ماشین-2021
The purpose of this work is aiming at the problem of high precision and large-scale measurement of motion parameters of aircraft ground test. Based on the traditional vision measurement method of motion parameters, a new measurement method of motion parameters based on intersected planes is discussed. Firstly, the measurement of 3D coordinates of laser spots was performed using the 3 D vision technique combined with plane constraints of curtain wall, the images of laser spot projected on the parallel curtain wall were processed using the centroid method, the calibration of multi-coordinates were unified by the unique coordinate system method, then the pose was solved using the proposed method that the normal vector of the intersect plane fixed on the vehicle is used. Finally, according to the measurement principle, the influence of spot center positioning, camera parameter calibration and target geometric error on the calculation results of aircraft motion parameters is analyzed, and the error propagation model is given. The experimental results show that in the measurement range of 8000 mm × 4000 mm × 4000 mm, the measurement error of attitude parameters is less than 0.14◦, and the measurement error of position parameters is less than 2 mm. Compared with binocular stereo vision, the accuracy of attitude angle measurement is improved by 100%. Therefore, the measurement method proposed in this paper can achieve high precision and large-scale measurement of the motion parameters of the test aircraft.
Keywords: Machine vision | Aircraft ground test | Intersected planes | Motion parameters | Muti-coordinate system calibration
A feasibility research on the application of machine vision technology in appearance quality inspection of Xuesaitong dropping pills
یک امکان سنجی در مورد استفاده از فناوری بینایی ماشین در بازرسی کیفیت ظاهر قرص های رهاسازی Xuesaitong-2021
Defect detection is a critical issue for the quality control of dropping pills, which is a special dosage form of traditional Chinese Medicine. Machine vision is a non-destructing testing technology and cost-effective with high accuracy that can be used to predict the detects of both interior and exterior of the sample by employing the camera. In this research, a machine vision system for inspecting quality of the Xuesaitong dropping pills (XDPs) that include non-spherical, abnormal sizes and colors was developed to evaluate the appearance quality of XDPs rapidly and accurately. Firstly, 270 images of XDPs containing qualified and three different types of defects were collected. Subsequently, the processing of the XDPs images were carried out. Finally, Three defecting categories classification models were developed and compared based on contour and color features. The experimental results showed that the Random Forest outperformed all the explored models and the classification accuracy for non-spherical, abnormal sizes and colors reached 98.52%, 100.00% and 100.00%, respectively. In summary, the method established in this research is scientific, reliable, fast and accurate, which has great application potential and can provide technical support for the automatic defect detection of dropping pills.© 2021 Elsevier B.V. All rights reserved.
Keywords: Machine vision | Xuesaitong dropping pills | Defect detection | Classification model | Random forest
A Machine Vision Based Automated Quality Control System for Product Dimensional Analysis
سیستم کنترل کیفیت خودکار مبتنی بر بینایی ماشین برای تجزیه و تحلیل ابعاد محصول-2021
Quality control (QC) in manufacturing processes is critical to ensuring consumers receive products with proper functionality and reliability. Faulty products can lead to additional costs for the manufacturer and damage trust in a brand. A growing trend in QC is the use of machine vision (MV) systems because of their noncontact inspection, high repeatability, and relatively low cost. This paper presents a robust MV system developed to perform comparative dimensional inspection on diversely shaped samples, including additive manufacturing products. The algorithm used performs dimensional inspection on a base product considered to have acceptable dimensions. The perimeter, area, rectangularity, and circularity of the base product are determined using blob analysis on a calibrated camera. These parameters are then used as the standard with which to judge additional products. Each product following is similarly inspected and compared to the base product parameters. A likeness score is calculated for each product, which provides a single value tracking all parameter differences. Finally, the likeness score is considered on whether it is within a threshold, and the product is considered to be acceptable or defective. The proposed MV system has achieved satisfactory results, as discussed in the results section, that would allow it to serve as a dependable and accurate QC inspection system in industrial settings.© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0)Peer-review under responsibility of the scientific committee of the Complex Adaptive Systems Conference, June 2021.
Keywords: Machine Vision | Quality Control | Dimensional Analysis | Digital Quality | Rectangularity | Circularity | Production | Manufacturing
Vision-based path detection of an automated guided vehicle using flower pollination algorithm
تشخیص مسیر مبتنی بر دید یک وسیله نقلیه هدایت شده خودکار با استفاده از الگوریتم گرده افشانی گل-2021
The automated guided vehicle (AGV) with sensor recognition method is having shortcomings such as high cost and noise. In this regard, the flower pollination algorithm (FPA) is applied in the path detection system of an AGV, in combination with computer vision. The path detection system starts with image acquisition using an onboard camera. The captured image is then preprocessed to obtain simple contrast of the path. Subsequently, the FPA is used to find a set of solution points fall inside the path zone. A regression model is formed as path guidance for AGV to travel accordingly. The performance of FPA is analyzed via simulation and real-world experiment using a robotic platform and tested on commonly seen line patterns in the industrial environment. The effectiveness of FPA in path detection is also com- pared with particle swarm optimization. The obtained results demonstrate the promising feasibility of the proposed FPA-based path detection system.© 2020 The Authors. Published by Elsevier B.V. on behalf of Faculty of Engineering, Ain Shams University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Automated guided vehicle | Flower pollination algorithm | Line detection | Line follower | Machine vision
Machine vision for drill string slip status detection
بینایی ماشین برای تشخیص وضعیت لغزش رشته مته-2021
Slip status of drill string is system generated binary value computed by comparison of sensor generated real time hook load value with a minimum threshold value of hook load stored in measurement system. This research article describes a novel method of slip status detection by machine vision technology which helps overcome the constraints of slip status detection with legacy measurement method. It also helps improve the real time drilling data quality and optimize and automate drilling operations. A method to detect drill string slip status with high-resolution digital camera installed on mast near rig ﬂoor is described along with backend vision processing and communication modules, which generate binary values of slip status. The binary values are transferred in real time to drilling measurement system of rig to compute other drilling parameters like bit depth, hole depth and stand counters. This method includes deploying active optical sensors at the rig ﬂoor, obtaining 1-D, 2-D, or 3D image data, and processing it to obtain the status of drill string. Reliable measurement of slip status by machine vision helps reduce non-productive time (NPT) by reliable real time surveillance of drilling operations.© 2021 Chinese Petroleum Society. Publishing services provided by Elsevier B.V. on behalf of KeAi Communication Co. Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Oil and gas | Machine vision | Camera | Slip-status | Rig | Drilling | Rotary table | NPT | Drilling fluid | Drill pipe
Plant trait estimation and classification studies in plant phenotyping using machine vision – A review
برآورد و طبقه بندی صفات گیاهی در فنوتیپ سازی گیاهان با استفاده از بینایی ماشین - مرور-2021
Today there is a rapid development taking place in phenotyping of plants using non-destructive image based machine vision techniques. Machine vision based plant phenotyping ranges from single plant trait estimation to broad assessment of crop canopy for thousands of plants in the field. Plant phenotyping systems either use single imaging method or integrative approach signifying simultaneous use of some of the imaging techniques like visible red, green and blue (RGB) imaging, thermal imaging, chlorophyll fluorescence imaging (CFIM), hyperspectral imaging, 3-dimensional (3-D) imaging or high resolution volumetric imaging. This paper provides an overview of imaging techniques and their applications in the field of plant phenotyping. This paper presents a comprehensive survey on recent machine vision methods for plant trait estimation and classification. In this paper, information about publicly available datasets is provided for uniform comparison among the state-of-the-art phenotyping methods. This paper also presents future research directions related to the use of deep learning based machine vision algorithms for structural (2-D and 3-D), physiological and temporal trait estimation, and classification studies in plants.
Keywords: Plant phenotyping | Machine vision | Plant trait estimation | Imaging techniques | Leaf segmentation and counting | Plant classification studies