Edge-Based Licence-Plate Template Matching for Identifying Similar Vehicles

This paper presents licence-plate recognition for identifying vehicles with similar licence-plates. The method uses a modified licence-plate recognition pipeline, with licence-plate template matching replacing character segmentation and recognition. Only edge detection is used, combined with a method for calculating line ratio to locate and extract licence-plates. The extracted licence-plate templates are then compared for licence-plate matching. The results show that the method performs well in differing circumstances, and that it is computationally cost-effective. Results also show that licence-plate template matching is a reliable method of identifying similar vehicles, and has a lower computational cost when compared with character segmentation and recognition.

Download Full-text

Proposal for Automatic Vehicle Number Plate Recognition System in POLIMAS

Journal of Engineering and Science Research ◽

10.26666/rmp.jesr.2020.1.4 ◽

2020 ◽

Vol 4 (1) ◽

pp. 20-27

Author(s):

Keyword(s):

Edge Detection ◽

Traffic Control ◽

Character Recognition ◽

Template Matching ◽

Optical Character Recognition ◽

Character Segmentation ◽

Identification System ◽

Recognition Mechanism ◽

Detection Techniques ◽

Plate Number

Automatic Vehicle Number Plate Recognition (AVNPR) system is an image processing technology in computer vision which captures the image of the vehicle and recognizes its number plate. Currently, the security and management of transportation system becomes an important key in controlled place such as campus area. With an increased number of vehicles, there is a need for vehicle recognition mechanism that is effective, affordable and efficient. The objective of this paper is to propose an efficient automatic vehicle identification system by recognizing the vehicle plate number. The system will be installed at the main entrance of POLIMAS to ensure that only the authorized vehicles can automatically enter the campus area. Once the vehicle is detected by the input sensors, AVNPR system will capture the image of vehicle plate number. An image is then extracted and investigated character segmentation by using optical character recognition (OCR). The method used for detection of a plate number is by pre-processing of the image and using a combination of Sobel edge detection and Laplacian edge detection techniques. Bounding Box technique is used to find the number plate and character segmentation. For character recognition, OCR method is used by using template matching to compare the segmented image with the template. The system is sustainable as the camera will only be switched on when a car is present. The propose system successfully detects and recognizes the vehicle number plate on real images. This system can also be used for security and traffic control.

Download Full-text

Automatic Receipt Recognition System Based on Artificial Intelligence Technology

Applied Sciences ◽

10.3390/app12020853 ◽

2022 ◽

Vol 12 (2) ◽

pp. 853

Author(s):

Cheng-Jian Lin ◽

Yu-Cheng Liu ◽

Chin-Ling Lee

Keyword(s):

Character Recognition ◽

Template Matching ◽

Recognition Accuracy ◽

Recognition System ◽

Character Segmentation ◽

Small Object ◽

Labor Costs ◽

Accuracy Rate ◽

Artificial Intelligence Technology ◽

S Model

In this study, an automatic receipt recognition system (ARRS) is developed. First, a receipt is scanned for conversion into a high-resolution image. Receipt characters are automatically placed into two categories according to the receipt characteristics: printed and handwritten characters. Images of receipts with these characters are preprocessed separately. For handwritten characters, template matching and the fixed features of the receipts are used for text positioning, and projection is applied for character segmentation. Finally, a convolutional neural network is used for character recognition. For printed characters, a modified You Only Look Once (version 4) model (YOLOv4-s) executes precise text positioning and character recognition. The proposed YOLOv4-s model reduces downsampling, thereby enhancing small-object recognition. Finally, the system produces recognition results in a tax declaration format, which can upload to a tax declaration system. Experimental results revealed that the recognition accuracy of the proposed system was 80.93% for handwritten characters. Moreover, the YOLOv4-s model had a 99.39% accuracy rate for printed characters; only 33 characters were misjudged. The recognition accuracy of the YOLOv4-s model was higher than that of the traditional YOLOv4 model by 20.57%. Therefore, the proposed ARRS can considerably improve the efficiency of tax declaration, reduce labor costs, and simplify operating procedures.

Download Full-text

A Cost-Effective Method for Bulge Prediction in Two Point Incremental Forming

10.25518/esaform21.419 ◽

2021 ◽

Author(s):

Jaekwang Shin ◽

Ankush Bansal ◽

Randy Cheng ◽

Alan Taub ◽

Mihaela Banu

Keyword(s):

Shape Function ◽

Incremental Forming ◽

Computational Cost ◽

Region Of Interest ◽

Computation Time ◽

Mesh Size ◽

Cost Effective ◽

Summation Method ◽

Fem Modeling ◽

Forming Defects

Accurate prediction of the defects occurring in incrementally formed parts has been gaining attention in recent years. This interest is because accurate predictions can overcome the limitation in the advancement of incremental forming in industrial-scale implementation, which has been held back by the increase in the cost and development time due to trial and error methods. The finite element method has been widely utilized to predict the defects in the formed part, e.g., bulge. However, the computation time of running these models and their mesh-size dependency in predicting the forming defects represent barriers in adopting these models as part of CAD-FEM-CAE platforms. Thus, robust analytical and data-driven algorithms must be developed for a cost-effective design of complex parts. In this paper, a new analytical model is proposed to predict the bulge location and geometry in two point incremental forming of an aerospace aluminum alloy AA7075-O for a 67° truncated cone. First, the algorithm calculates the region of interest based on the part geometry. A novel shape function and weighted summation method are then utilized to calculate the amplitude of the instability produced by material accumulation during forming, leading to a bulge on the unformed portion of the sample. It was found that the geometric profile of the part influences the shape function, which is a function created to incorporate the effects of process parameter and boundary condition. The calculated profile in each direction is finalized into one 3-dimensional profile, compared with the experimental results for validation. The proposed model has proven to predict an accurate bulge profile with 95% accuracy comparing with experiments with less than 5% computational cost of FEM modeling.

Download Full-text

A Vision-Based Approach for Sidewalk and Walkway Trip Hazards Assessment

International Journal of Environmental Research and Public Health ◽

10.3390/ijerph17228438 ◽

2020 ◽

Vol 17 (22) ◽

pp. 8438

Author(s):

Rachel Cohen ◽

Geoff Fernie ◽

Atena Roshan Fekr

Keyword(s):

Environmental Factors ◽

Edge Detection ◽

Gaussian Process ◽

Regression Models ◽

Gaussian Process Regression ◽

Region Growing ◽

Cost Effective ◽

Second Phase ◽

Cost Effective Method ◽

Regression Techniques

Tripping hazards on the sidewalk cause many falls annually, and the inspection and repair of these hazards cost cities millions of dollars. Currently, there is not an efficient and cost-effective method to monitor the sidewalk to identify any possible tripping hazards. In this paper, a new portable device is proposed using an Intel RealSense D415 RGB-D camera to monitor the sidewalks, detect the hazards, and extract relevant features of the hazards. This paper first analyzes the effects of environmental factors contributing to the device’s error and compares different regression techniques to calibrate the camera. The Gaussian Process Regression models yielded the most accurate predictions with less than 0.09 mm Mean Absolute Errors (MAEs). In the second phase, a novel segmentation algorithm is proposed that combines the edge detection and region-growing techniques to detect the true tripping hazards. Different examples are provided to visualize the output results of the proposed method.

Download Full-text

Surface-offset gathers from elastic reverse time migration and velocity analysis

Geophysics ◽

10.1190/geo2018-0676.1 ◽

2020 ◽

Vol 85 (1) ◽

pp. S47-S64

Author(s):

Yang Zhao ◽

Tao Liu ◽

Xueyi Jia ◽

Hongwei Liu ◽

Zhiguang Xue ◽

...

Keyword(s):

Velocity Ratio ◽

Computational Cost ◽

Cost Effective ◽

Velocity Estimation ◽

Reverse Time ◽

Reverse Time Migration ◽

S Wave ◽

Time Migration ◽

The Common ◽

Wave Decomposition

Angle-domain common-image gathers (ADCIGs) from elastic reverse time migration (ERTM) are valuable tools for seismic elastic velocity estimation. Traditional ADCIGs are based on the concept of common-offset domains, but common-shot domain implementations are often favored for computational cost considerations. Surface-offset gathers (SOGs) built from common-offset migration may serve as an alternative to the common-shot ADCIGs. We have developed a theoretical kinematic framework between these two domains, and we determined that the common SOG gives an alternative measurement of kinematic correctness in the presence of incorrect velocity. Specifically, we exploit analytical expressions for the image misposition between these two domains, with respect to the traveltime perturbation caused by velocity errors. Four formulations of the PP and PS residual moveout functions are derived and provide insightful information of the velocity error, angle, and PS velocity ratio contained in ERTM gathers. The analytical solutions are validated with homogeneous examples with a series of varied parameters. We found that the SOGs may perform in the way of simplicity and linearity as an alternative to the common-shot migration. To make a full comparison with ADCIGs, we have developed a cost-effective workflow of ERTM SOGs. A fast vector P- and S-wave decomposition can be obtained via spatial gradients at selected time steps. A selected ERTM imaging condition is then modified in which the migration is done by offset groups between each source and receiver pair for each P- and S-wave decomposition. Two synthetic (marine and land) examples are used to demonstrate the feasibility of our methods.

Download Full-text

A Cost-Effective and Robust Edge-Based Blur Metric Based on Careful Computation of Edge Slope

IEICE Transactions on Information and Systems ◽

10.1587/transinf.e94.d.1834 ◽

2011 ◽

Vol E94-D (9) ◽

pp. 1834-1838 ◽

Cited By ~ 1

Author(s):

Hanhoon PARK ◽

Hideki MITSUMINE ◽

Mahito FUJII

Keyword(s):

Cost Effective ◽

Edge Based

Download Full-text

A Quasi-Generalized-Coordinate Approach for Numerical Simulation of Complex Flows

Journal of Fluids Engineering ◽

10.1115/1.2354533 ◽

2006 ◽

Vol 128 (6) ◽

pp. 1394-1399 ◽

Cited By ~ 1

Author(s):

Donghyun You ◽

Meng Wang ◽

Rajat Mittal ◽

Parviz Moin

Keyword(s):

Coordinate System ◽

Stokes Equations ◽

Computational Cost ◽

Three Dimensional ◽

Cost Effective ◽

Navier Stokes ◽

Two Dimensional ◽

Navier Stokes Equations ◽

Curvilinear Coordinate ◽

Generalized Coordinate

A novel structured grid approach which provides an efficient way of treating a class of complex geometries is proposed. The incompressible Navier-Stokes equations are formulated in a two-dimensional, generalized curvilinear coordinate system complemented by a third quasi-curvilinear coordinate. By keeping all two-dimensional planes defined by constant third coordinate values parallel to one another, the proposed approach significantly reduces the memory requirement in fully three-dimensional geometries, and makes the computation more cost effective. The formulation can be easily adapted to an existing flow solver based on a two-dimensional generalized coordinate system coupled with a Cartesian third direction, with only a small increase in computational cost. The feasibility and efficiency of the present method have been assessed in a simulation of flow over a tapered cylinder.

Download Full-text

Research and Implementation of Vehicle-Logo Recognition Based on Modified Invariant Moments

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.717.444 ◽

2013 ◽

Vol 717 ◽

pp. 444-448 ◽

Cited By ~ 2

Author(s):

Min Huang ◽

Guo Feng Yang ◽

Ya Qiong Ma ◽

Yan Ming Wang

Keyword(s):

Edge Detection ◽

Template Matching ◽

Minimum Distance ◽

Recognition Rate ◽

Experimental Results ◽

Invariant Moments ◽

Logo Recognition ◽

Characteristic Value ◽

Vehicle Logo Recognition

According to the low recognition rate of Hu invariant moments in the target images, this article proposes a vehicle-logo recognition research algorithm based on the modified invariant moments. At first, use the template matching to locate the vehicle-logo rough area and use the edge detection for accurate location. Then, calculate the characteristic value of the modified invariant moments of the vehicle-logo, finally, the vehicle-logo is recognized according to the minimum distance of invariant moments. The experimental results show that the modified invariant moments can improve the recognition rate of target images effectively.

Download Full-text