Man-to-man vehicle monitoring is the existing vehicle license plate recognition in the province of Eastern Samar, specifically in Eastern Samar State University (ESSU), the manual monitoring of vehicle makes a problem, due to the ratio of vehicle to worker. Hence, inaccurate and inefficient monitoring is the consequence. As the mobility goes on, the main objective of the study is to develop a software that will serve as the initial replacement to a man-to-man vehicle license plate status recognition, primarily applied to motorbike vehicle, a solution to inefficient and inaccurate motorbike monitoring. Developing a proposed computerized system is an initial solution and the best solution to the problem.
In our country, the Philippines, 95% replaced the logbook mode to biometrics machine in log-in and log-out recording of any government staff and personnel. Biometrics machine uses image processing in individual identification and database storage.
The main concern of any image processing mechanism and transaction is the security. The security is negotiable in accordance to software features and limitations. In this system, limitations are characteristics of License Plate (LP) which are non-recognizable aspects, and they are excluded, such as: (1) the physical aspect of the license plate if it is faded and damaged; (2) do-it-yourself (DIY) plate; and (3) license plate character is overlapped by a text, markers, or stickers. Actually, more image processing mode of transaction are being practice in local and national agencies. Security services office of ESSU with the coordination of LTO are partners in the development of the system. A close-circuit television (CCTV) camera was installed in the ESSU vehicle entrance, and available in capturing, recognizing, and recording the motorbike license plate. Hence, the study is allied concern of two government entity, the Land Transportation Office (LTO), and the Higher Education Institution (HEI), ESSU in particular.
Plate detection and recognition of Iraq License Plate Using KNN algorithm, an automated car license plate recognition system for Iraq vehicle plate number which is developed to control and apply the law enforcement related. It processed the license plate recognition in three stages; preprocessing, license plate localization, license plate recognition. The license plate image is preprocessed through converting the image to grayscale and apply morphological transformation filter to convert the result to binary image. Then, distort the binary image using Gaussian filter, find the contours in the image using OpenCV (Open Source Computer Vision) Library
Automated Vehicle License Plate Detection using KNN as a method, KNN identifies the license plate accurately. Preprocessing was the initial step with the aid of median filtering approach. After preprocessing, the system extracts the license plate image of characters. The license numbers were recognized from the extracted license plate with the help of character segmentation approach. Then it was accurately recognized using the machine learning technique which is the KNN classifier
Addresses the problem of vehicle plate number localization from simple and complex backgrounds using a novel approach which is based on a modified GrabCut algorithm as an automatic localization technique for identification and extraction of LPs from captured vehicle images
In this proposal, the license plate localization, the KNN (K-nearest neighbors) algorithm was used in finding possible characters in the image. Cropping the part of the image with highest candidate license plate, and apply preprocessing, localization, and recognizing all part of the license plate in the cropped image. Python was used in coding the program that makes the system friendly and simple in coding for C++ user, than mathlab, and the database is SQLite, with the integration of these different software makes the system unique and distinct.
As the system developed, evaluation was conducted with these components: (1) functionality; (2) efficiency; (3) accuracy. Adopted on International Standardization Organization (ISO) 9126. Since this system is exclusively used for ESSU motorbike entrants.
The average motorbike entrants in every twenty-four hours are 480-motorbikes in approximate population of ESSU civilians of 4,680. Majority are commuters, other vehicle are non-motorbikes. Why motorbikes? 85% vehicle entering the campus are motorbikes, because of accessibility. Vehicle with high violations are motorbikes, somebody parts of the motorbikes and the scooters are not functioning like signal lights, horns which is dangerous to operators and individuals who pass the same road.
Despite the success of license plate detection (LDP) method in the past decades, only a few methods can effectively detect multi-style license plates (LPs), especially from different countries
System development and evaluation was used which focused on innovation and strategic learning rather than standard outcomes.
Study was conducted in ESSU, Main Campus, Borongan City, in the Province of Eastern Samar, Philippines.
The software development model used was the Waterfall model. This model is commonly used for small or mid-size projects with clearly defined and unchanging requirements. This software model will help the researchers to manage each procedure during the development process of the proposed system.
1. Requirement Gathering -Includes defining the needs for the development of the proposal and also specifying the limits and scope of the study.
2. Design - This phase includes designing the system features along with its proposed user interface.
3. Implementation -The coding takes place. Defines the implementation of the proposed designs and features to the actual system. This will be done with the use of software/tools selected for the development.
4. Testing - During this stage, an initial version of the system is done and ready to be tested by the end user.
5. Deployment -The system will be distributed to end users for evaluation of the acceptance in terms of its functionality, efficiency, and accuracy.
6. Maintenance - In this phase, it defines the tweaks needed for the errors encountered by the end user for the system to improve its functions.
Python 3.6 was used in program coding, the Graphical User Interface (GUI) used was Tkinter, the database was SQLite, and in image processing OpenCV was utilized.
|
|
Processor |
Processor: Intel Celeron CPU: N3450 @ 1.10 GHZ |
RAM |
4.00GB(3.83 GB usable) |
System Type(OS) |
64-bit OS x 64-based processor |
Using Slovin’s formula was used in finding the sample of respondents. The population size applied is 22 security guards from Security Services Office, and 14 faculty members in Information Technology Education ITE department, having the population of 36. Using the margin of error is 5%, and the confidence level is 95%, and the required sample size is 33 respondents.
Unstructured survey questionnaire was used. The ISO 9126 Quality Software used as bases for the software evaluation. It includes the functionality parameter, by identifying both internal and external quality characteristics of software product. It has three representations for defining and identifying the quality of software product; functionality, efficiency, accuracy.
Frequency distribution, percentage, and weighted mean were used to interpret the data gathered from the software evaluation metrics. Likert scale was used as measurement scale for the different software criteria, and weighted mean to determine the actual unified evaluation result.
Likert Scale System
Scale Interpretation
5 Strongly Agree
4 Agree
3 Neutral
2 Disagree
1 Strongly Disagree
|
|
4.2 – 5.0 |
Highly Acceptable |
3.4 - 4.1 |
Acceptable |
2.6 – 3.3 |
Neither Acceptable nor Unacceptable |
1.8 – 2.5 |
Unacceptable |
1 – 1.7 |
Highly Unacceptable |
The Graphical User interface (GUI) of the initial entry is in
The respondent’s extent of agreement in terms of functionality of the system is illustrated in
Functionality |
Percentage |
|||||||
Yes |
No |
|||||||
The system can recognize the plate number of the motorbike. |
100% |
0% |
||||||
The system can identify registered motorbike. |
100% |
0% |
||||||
The system has all functions and capability that was expected to have. |
100% |
0% |
||||||
Efficiency |
Response |
Extent of Agreement |
||||||
5 |
4 |
3 |
2 |
1 |
WM |
Description |
Interpretation |
|
The systems provide appropriate response and processing times and throughput rates when performing its function under stated conditions. |
11 |
16 |
6 |
0 |
0 |
4.15 |
Strongly Agree |
Acceptable |
The system used appropriate amounts and types of resources when the software performs its function under stated condition. |
15 |
15 |
3 |
0 |
0 |
4.36 |
Strongly Agree |
Highly Acceptable |
Accuracy |
Response |
Extent of Agreement |
||||||
5 |
4 |
3 |
2 |
1 |
WM |
Description |
Interpretation |
|
The system validates input of registered user. |
28 |
5 |
0 |
0 |
0 |
4.84 |
Strongly Agree |
Acceptance test |
The system is precise in its results in terms of tracking documents. |
23 |
8 |
2 |
0 |
0 |
4.63 |
Strongly Agree |
Highly Acceptable |
|
|
|
Efficiency |
4.25 |
Highly Acceptable |
Accuracy |
4.73 |
Highly Acceptable |
Overall Mean |
4.49 |
Highly Acceptable |
Data presented tune-in the objectives of the study, the status of the motorbike was efficiently recognized which could be a big help to LTO, the huge agency that apprehend those violators, and monitor the possible number of motorbikes officially registered. ESSU Security Services Office can replace the man-to-man monitoring of motorbikes coming-in and going-out of the Campus, less effort and less time consuming.
In comparison of the study made by
The proposal replaced the man-to-man motorbike license plate monitoring. The system is functional, efficient, and accurate in recognizing and showing the status of the motorbike. From the evaluation, the overall acceptance of the software is highly acceptable, hence can be utilized for e-recognition of motorbike LP. Methods applied in software development can be expanded for large number of motorbike or even to include other types of vehicles.
Limitations or non-recognizable pieces found in the license plate are recommended to be detected for further study, features such as: faded and damaged plate, do-it-yourself (DIY) plate, and license plate character is overlapped by a text, markers, or stickers.