Defected Tablet Detection System

Project Title

Defected Tablet Detection System

Project Area of Specialization Software EngineeringProject Summary

The world is moving so fast for the technology usage and computer vision technology. Computer vision plays a great role in the medicine field. The summary of the project is that we have converted the manual inspection of medicines with the automated inspection through computer vision in order to provide efficient, effective, hygienic  and less time consumption way of inspection. Furthermore, we have provided a report analysis feature through which company’s manager can analyze the defect rate of the medicine and can manage the error rate and employees details. The proposed solution can detect the shape, crack, scratch, missing, color and broken tablets. We have proposed a defect detection system  for the pharmaceutical companies  and compared our technique with the existing defect detection method. We also have shown a comparison for defect detection and processing time between the proposed method and the existing method. Detection rate of the proposed method is better than that of the existing method and we need less time to detect defects than the existing one.  The world is moving so fast for the technology usage and computer vision technology. Computer vision plays a great role in the medicine field. The summary of the project is that we have converted the manual inspection of medicines with the automated inspection through computer vision in order to provide efficient, effective, hygienic  and less time consumption way of inspection. Furthermore, we have provided a report analysis feature through which company’s manager can analyze the defect rate of the medicine and can manage the error rate and employees details. The proposed solution can detect the shape, crack, scratch, missing, color and broken tablets. We have proposed a defect detection system  for the pharmaceutical companies  and compared our technique with the existing defect detection method. We also have shown a comparison for defect detection and processing time between the proposed method and the existing method. Detection rate of the proposed method is better than that of the existing method and we need less time to detect defects than the existing one. 

Project Objectives

The main objective of this system is to use  image processing, which is better than x-ray scanning and human inspection, image processing is a method to perform some operations on an image in order to get an enhanced image or to extract some useful information from it. It is a type of signal in which an input of an image and the output is image characteristics or features associated with that image. Our category is tablet detection, tablets detection are divide into many characteristic and parameter to detect it faults. The training phase obtains the image of the tablet blister and stores it characteristics such as shape of tablets, color of the tablets, size, and the number of the tablets. The testing phase consists of obtaining the tablet's image from the manufacturing line and comparing it with the golden image, which is (defectless). The first image will be converted into greyscale and then to binary further noise is remove for tablets shaped. To identify the damaged tablets after production involves a series of steps involving image enhancement, segmentation, thresholding, filtration, pixel calculation, elimination of noise and region-based static to identify the broken tablets.

Project Implementation Method

The implementation method consists the following steps:

User: User can select the tablets from the screen, can view the process, are notify by the system.

Input: System will have database that will be having correct images, so that system can compare the images with the current image, system will only check for the limited type of tablets. We can resolve that issues in our project with MATLAB tool using digital image processing algorithm, we are using any high graphical camera, in which we can set programming fundamentals to optimize tablets production we combine many factors to make that process more efficient. The train image will be preprocessed, stored in database and test image containing various defects were processed, compare to eliminate the tablets from the manufacturing using MATLAB/image processing algorithm and the defected result will be show in the screen.

Process: This process consists of identifying the various characteristics of the tablets such as color, shape, size and number of tablets in the blister. With the binary image of the blister, the shape is characterized as circle, rectangle, or arbitrary along with checking the blister is empty or not but it’s requires cameras of high resolutions for all the images detection and then compare with the perfect image which is defect less (trained model) in which include complex coding. For processing of task and scope of project would be the user control in the application how he/she is interacting to the application performing and managing task.

Output: System will generate the statistical report; statistical report will show the accuracy rate and the number of tablets being inspect either right or wrong.

Defects: System will detect the defects like (size, shape, color, positions and make

counts) of the tablets.

Benefits of the Project

Digital data of the medicines is not changeable and always retains its? ?originality; irrespective of how many times the data? ?is reproduced.? ?Digital image processing is a powerful tool for the medicine inspectors ? ?that moderate the search for representative defects.? ?Once ? ?the ? ?image ? ?is ? ?acquired ? ?then ? ?immediately ? ?it? ?displays.? ?It quantifying the changes over time.? ?Quick comparisons of images can be done, so that it would save time of inspection.?

Technical Details of Final Deliverable

• Project Deliverable 1: The system will detect the defaults like shape, size, and position of the image by comparing it with the correct image.
• Project Deliverable 2: We will enhance algorithm’s condition according to our project requirements and its efficiency for time.
• Product Deliverable 3: System will generate statistical report so that admin could make analysis.

• Defining the technical details of the final deliverables through risk plan, where levels:

  H = High

  M = Medium

  L = Low

  Risk Area	Level (H/M/L)	Risk Plan
  Project risk #1: Capturing clear image and applying pre-processing.	M	Risk plan activity #1: Using suitable image pre-processing technique. Risk plan activity #2: Using better quality camera
  Project risk #2: Items are placed properly according to their positions.	M	Risk plan activity #1: Mark their spots on conveyor belt.
  Project risk #3: Accuracy	H	Risk plan activity #1:Using good algorithm Risk plan activity #2: Proper training of the system.
  Project risk #4:Time Complexity	H	Risk plan activity #1:Better system and hardware

Risk Area

Project risk #1: Capturing clear image and applying pre-processing.

Project risk #2: Items are placed properly according to their positions.

Project risk #3: Accuracy

Project risk #4:Time Complexity

Final Deliverable of the Project Software SystemCore Industry MedicalOther Industries IT Core Technology OthersOther Technologies OthersSustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Elapsed time in (days or weeks or month or quarter) since start of the project	Milestone	Deliverable
Month 1	Milestone 1 -	Initial ProposalSystem diagramDetailed ProposalNBP competition ProposalFeature List
Month 2	Milestone 2 -	Based on each use caseClass DiagramFront-end (GUI/Web/ Mobile)Sequence DiagramObject Diagram and EntityRelation Diagram
Month 3	Milestone 3 -	Working project in executable form with component anddeployment diagrams.? Deployment Architecture andStrategy