Collect
Collect representative product and defect images using the inspection equipment.
Precision Inspection · Confidence in Every ProductEnglish
AI Deep Learning Platform
AI Inspection for Capsules, Tablets, Softgels and Blister Packs
Develop inspection models for pharmaceutical and nutraceutical capsules, tablets, softgels and blister packs. Explore product-specific examples and connect recipes and models with AIACVISION inspection equipment.
Experience AI
Experience AI
Select a product and an example, then click AI Inspection to view the corresponding result.
Original Image
AI Inspection Result
Click AI Inspection to view the result.
Build a model for your product
From sample images
to machine deployment.
The AIACVISION Online AI Deep Learning Platform supports a guided workflow for developing models for new products and new defect types.
Annotate
Upload images to the platform, choose defect categories and label the relevant regions.
Train
Train a model using predefined training settings.
Validate
Evaluate detection and false rejects with representative samples kept separate from the training set before applying the model to production.
Deploy
Download and deploy the model and machine parameters to compatible AIACVISION equipment.
Continuous improvement
New products.
New defects.
Keep improving.
Extend your defect library
Add new defect samples and annotations as product requirements evolve.
Update and evaluate models
Retrain with the expanded dataset, evaluate detection and false rejects, and review deployment through your change-control process.
A guided training workflow
Predefined settings and a structured labeling process simplify model development.
Online Recipe Setup and Training
Online Recipe Setup and Training
Build practical recipe-development skills before applying settings to production equipment.
Practice with Product and Defect Images
The online training system includes acceptable and defective product images across multiple product types. Select relevant examples to practice defect definition, Pattern Recognition settings, AI model preparation and performance evaluation.
Guided Instruction and Application Support
Targeted exercises support operators, quality personnel and technical teams. AIACVISION can assist with sample selection, labeling, training and performance review, with sessions arranged for the installed equipment and products.
Online Recipe Training is a guided training service. The public Experience AI section continues to show selected examples; it does not provide a production-connected recipe editor.
Built for AIACVISION equipment
Connect AI models
with your inspection solution.
Capsule Inspection
AICIM capsule vision inspection machines
View SolutionSolid Dose Inspection
AISIM tablet, capsule and softgel inspection
View SolutionBlister Inspection
AIBIM product, packaging and printing inspection
View SolutionCounting-Line Inspection
Vision modules for slat and electronic counters
Find Your SolutionCapsule Inspection Examples
Explore opaque, transparent and liquid-filled capsule examples. Product appearance and defect types determine lighting, recipes and model requirements.
Tablet Vision Inspection and Softgel Vision Inspection
Use the examples to review product-specific defects. Discuss actual good and defective samples when evaluating a model for production.
Friendly Recipe Setup and Management
Product recipes and inspection models are prepared for the selected product. Changes to products or defect criteria require review and evaluation before production use.
Questions About AI Inspection for Capsules, Tablets, Softgels and Blister Packs
How are inspection recipes set up and managed?
Recipes and models are prepared for the product and defect criteria. Review changes with representative samples before deployment.
Can the system evaluate my product and defect samples?
Contact the team with your product and defect information to arrange an evaluation.
How are AI inspection results reviewed?
Review the visible example results, then use representative samples to assess the recipe against actual inspection requirements.
How should new AI inspection models be evaluated?
Evaluate detection and false rejects with representative acceptable products and known defects kept separate from the training set. Review performance before production use.
How are model updates reviewed before deployment?
Add relevant defect examples, retrain and evaluate the updated model. Review deployment through the customer change-control process, with application support available from AIACVISION.
Let's discuss your application
Let’s evaluate
your inspection requirements.
Tell us your product format, target defects and required capacity. We will help you evaluate the next step.