Automatic Paper Marking System
Nowadays, Multiple Choice Questions (MCQs) are inevitable part of almost all examinations. Even the admission tests of different schools, colleges, and universities a significant part of the question paper comprises of the MCQs. Checking the MCQs manually would have been a tiring and troublesome tas
2025-06-28 16:25:27 - Adil Khan
Automatic Paper Marking System
Project Area of Specialization Artificial IntelligenceProject SummaryNowadays, Multiple Choice Questions (MCQs) are inevitable part of almost all examinations. Even the admission tests of different schools, colleges, and universities a significant part of the question paper comprises of the MCQs. Checking the MCQs manually would have been a tiring and troublesome task for human-beings. Therefore, Optical Mark Recognition (OMR) sheets were invented. The OMR can be scanned and read by an OMR machine to find out the marks obtained by the students. The OMR is being extensively used by various organizations, such as National Testing Service (NTS), Educational Testing & Evaluation Agency (ETEA), National Medical & Dental College Admission Test (NMCAT), and Graduate Assessment Test (GAT) etc. for marking OMR sheets. Nowadays, to mark the papers automatically, there are two methods mainly used, which are, (i) the OMR sheet is scanned and marked through predefined automated mechanism [1], [2] and (ii) mobile applications, which capture the image of the OMR sheet and then mark them [3].
Both methods are reliable and accurate. However, the OMR machine available in market are based on scanning technology that is the OMR sheet is scanned and then further processed to mark the OMR sheets. The cost of those OMR machines depends upon various factors, such as manufacturing brand, reliability, size, and the most importantly how many sheets the OMR machine can mark in one minute. Whilst this project will focus to develop an automatic paper marking machine using Raspberry-Pi, which will capture an image of the OMR sheet placed under Raspberry-Pi camera and the image is read, aligned, marked and saved in a database along with information, such as name of student, registration number, total marks, and obtained marks. Specifically, the proposed project aims to achieve high accuracy and robustness, in real-time.
Project ObjectivesThe aim of the project is to develop a user friendly, cost efficient, robust, mobile, and accurate paper marking machine using Raspberry-Pi. Specifically main objectives of the proposed project are as following.
- We aim to develop a state-of-the-art paper marking system by developing novel algorithm. In addition, we also aim to develop a dedicated hardware setup, which will mark the OMR sheets automatically.
- We aim to save the information read from the OMR sheet in our database. For instance, we will use an excel sheet as database, where we will keep record of each OMR sheet along with important parameters, such as name, registration number, total marks, and obtained marks when once placed in the system.
- We focus to design a low-cost, simple, mobile, and accurate paper marking machine. Moreover, we intend to develop a system, which will be easy to use for non-technical staff.
To begin this project, we will start from understanding the developing and working of image processing algorithms and will explore different libraries of Image processing, which could be possibly used in our algorithm. Along with it, we also will explore the feasibility of Raspberry-Pi in depth. The schematic design of the proposed algorithm is depicted in Fig.1. After developing the algorithm, we will interface Raspberry-Pi camera and will perform testing of the algorithm. Then a hardware structure of machine will be designed and implemented which most likely will consist of a cubical shape with a tray on one side. We will place the OMR sheets inside our OMR machine and after marking it will roll to the other side of machine where the tray is installed. The supervisor and the project teammates are competent enough to perform all these tasks efficiently [4]-[8].
References
[1]. S. Hussmann and P. W. Deng, “A High-Speed Optical Mark Reader Hardware Implementation at Low Cost using Programmable Logic,” Real-Time Imaging, Vol. 11, No. 1, 2005, pp. 19–30.
[2]. A. Abbas, "An Automatic System to Grade Multiple Choice Question Paper Based Exams," Journal of Al-Anbar University for Pure Science, Vol. 3, No. 1, 2009, pp. 1–8.
[3]. A. A. Marakeby, “Multi-Core Processors for Camera based OMR,” International Journal of Computer Applications, Vol. 68, No. 13, 2013, pp. 1–5.
[4]. Z. Mahmood, K. Khan, U. Khan, S. H. Adil, S. S. A. Ali, and M. Shahzad, “Towards Automatic License Detection,” Sensors, Vol. 22, No. 3, 2022, pp. 1–19.
[5]. S. N. Khan, K. Khan, N. Muhammad, and Z. Mahmood, “Efficient Prediction Mode Decisions for Low Complexity MV-HEVC,” IEEE Access, 2021, pp. 150234–150251.
[6]. K. Khan, A. Imran, H. Z. Rehman, A. Fazil, M. Zakwan, and Z. Mahmood, “Performance Enhancement Method for Multiple License Plate Recognition in Challenging Environments,” EURASIP Journal on Image and Video Processing, No. 30, 2021, pp. 1–23.
Benefits of the ProjectDue to the development of the proposed project, we have the following benefits.
- It will be a cost efficient, robust, reliable, accurate and mobile OMR machine, which will save the unnecessary and time-consuming process to mark the MCQs based examination.
- It will provide fast and non-intrusive paper marking system. In addition, the proposed project will reduce the margin of errors whilst marking the OMR sheets.
- The proposed project will provide high order thinking, decision power, stability, availability, and security to the system.
- We believe that the proposed project is proactive, responsive, and works in real-time environment. Therefore, a newly hired staff with less experience can operate this product. Importantly, there is no need to train the personnel for marking OMR sheets.
References
[7]. N. Muhammad, N. Bibi, M. A. Shah, S. Zainab, I. Ullah, and Z. Mahmood,“An Entropy based Salient Edge Enhancement using Fusion Process,” Applied Mathematical Modelling, Vol. 93, 2021, pp. 426–442.
[8]. Z. Mahmood, N. Bibi, M. Usman, U. Khan, and N. Muhammad, “Mobile Cloud based Framework for Sports Applications,” Multidimensional Systems and Signal Processing, Vol. 30, No. 4, 2019, pp. 1991?2019.
[9]. Ansari, M. Aquib, D. Kurchaniya, and M. Dixit. "A comprehensive analysis of image edge detection techniques," International Journal of Multimedia and Ubiquitous Engineering Vol. 12, No.11, 2017, pp. 1?12.
[10]. C. Wang, “Fast Method for Rectangle Detection,” International Conference on Machinery, Materials, Environment, Biotechnology and Computer, Vol. 6, 2016, pp. 864-867.
[11]. Chiang, M. Chao, and T. E. Boult, “The integrating resampler and efficient image warping,” Proceedings of the ARPA Image Understanding Workshop, 1996, pp. 843–849.
[12]. Rahman, Ziaur, Y. F. Pu, M. Amir, and F. Ullah, “A framework for fast automatic image cropping based on deep saliency map detection and gaussian filter,” International Journal of Computers and Applications, Vol. 41, No. 3, 2019, pp. 207-217.
Technical Details of Final DeliverableThe flow of the proposed automatic paper marking system is depicted in Fig. 1. The system will comprise of the following interconnected steps:
Step-1: Entering the Parameters: The very first step after placing the OMR sheet in our paper marking system is to enter the parameters of the sheet, which involves following.
- No of questions and choices
- Correct Option
- Department and Class
Step-2: Image Acquisition: Once entered the aforedescribed parameters, the next step in expert systems involves the acquisition of the OMR sheet. When the OMR sheet is placed in our OMR machine, it will automatically be detected, and an image of OMR sheet is acquired either by an 8 mega-pixels camera embedded in Raspberry-pi.
Step-3: Pre-Processing: After image is captured, appropriate pre-processing techniques will be applied, which involves following steps.
- RGB to Double to Gray Conversion
- Binary Conversion
- Applying Gaussian Blur to smooth the Image
- Finding contours of image
- Applying Canny Edge Detector to highlight the contours of image [9]
Step-4: Warping, Cropping and Further Processing: On bases of contours of captured image, warping is applied, which is the process of digitally manipulating an image such that any shapes portrayed in the image have been significantly distorted. Later, the warped image is cropped and further processed, which involves following steps.
- Rectangle detection [10]
- Cropping the rectangles from warped Image [11]
- After successfully detecting the corners, those rectangles are further cropped and applied further processing [12].
Step-5: Post-Processing: On those portions, the MCQs marking, and reg no. reading algorithm will be applied. Phenomenon involved in marking MCQs is based on Pixel counts and comparison to the ground truth, which in our case is entered answer key in the system.
Step-6: Save and Display Results: After completion of above steps, the marks obtained are printed on the OMR sheet and saved in the system, while all information including name, class, department, reg no. is stored in Excel sheet and displayed on LCD attached to the system.
Step-7: Rolling Sheet to the tray: Once the whole process is done, sheet is rolled to a tray installed on the other side of machine. This occurs due to the roller installed along stepper motor, which is programmed to turn on for a specific instance of time when once the OMR sheet is marked and data is saved. This whole process will repeat itself, whenever an OMR sheet is detected.
Fig. 1: Architecture of proposed Automatic Paper Marking Machine
Final Deliverable of the Project HW/SW integrated systemCore Industry EducationOther Industries IT Core Technology Artificial Intelligence(AI)Other Technologies RoboticsSustainable Development Goals Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 72850 | |||
| Raspberry Pi-4 Model b with 8GB Memory | Equipment | 1 | 22000 | 22000 |
| Raspberry Pi Camera Module V2 8 Mega-pixels | Equipment | 1 | 5500 | 5500 |
| Raspberry Pi Official 7-Inch Capacitive Touch LCD Screen | Equipment | 1 | 14600 | 14600 |
| DC Power Supply for Raspberry Pi 4, 5V 5A with USB Type C cable | Equipment | 1 | 1500 | 1500 |
| Micro HDMI to female HDMI converter | Equipment | 1 | 250 | 250 |
| San Disk Ultra Micro SD card 32 GB Class (10) | Equipment | 1 | 1500 | 1500 |
| Jumper Wires male to male | Equipment | 1 | 400 | 400 |
| Jumper Wires male to female | Equipment | 1 | 400 | 400 |
| Paper Slide Roller | Equipment | 1 | 5800 | 5800 |
| KH56KM2U071 Printer Servo Stepper Motor | Equipment | 1 | 5700 | 5700 |
| Carbon Fiber Box (16’'x15’'x10'') | Equipment | 1 | 3500 | 3500 |
| Casing Box for Raspberry-Pi | Equipment | 1 | 700 | 700 |
| Casing Box for LCD Panel | Equipment | 1 | 1000 | 1000 |
| IEEE Conference Publication Cost | Miscellaneous | 1 | 7000 | 7000 |
| Thesis Costs | Miscellaneous | 3 | 1000 | 3000 |