# How a Robot System Inspects and Grades Objects for Flaws > This patent describes an automated system that uses cameras, special lighting, and deep learning AI to find and measure tiny flaws on an object's surface, then assigns a quality grade and sorts it. - **Patent:** US 10753882 - **Original title:** Inspection and cosmetic grading through image processing system and method - **Owner:** Griffyn Robotech Pvt - **Granted:** 2020 - **Status:** Active - **Times cited:** 25 - **Field:** consumer_electronics, automotive, semiconductors, software, ai_ml, mechanical, telecommunications ## What it does The system first uses an "image capture subsystem" to take pictures of an object, adjusting a "lighting subsystem" for optimal viewing conditions (Claim 1). An "image processing module" then employs "deep learning algorithms" to detect, segment, and classify surface defects, also measuring their length and width. For each identified defect, an "optical spot sensor subsystem" precisely measures its depth or height, providing a 3D understanding of the flaw (Claim 1). Finally, a "cosmetic grading module" determines the object's quality grade based on the number, type, and severity (length, width, and depth) of defects, and directs the item to the appropriate bin for shipment (Claim 3). For example, it could automatically inspect newly manufactured smartphone screens for scratches or dust particles. ## What it does NOT cover - Does not cover systems that rely solely on human visual inspection for grading objects. - Does not cover systems that only detect 2D surface defects without measuring their 3D depth or protrusion (Claim 1). - Does not cover systems that grade objects without using deep learning algorithms for defect detection and classification (Claim 1, 2). - Does not cover systems that do not use an automated control system to align the object for 3D defect measurement (Claim 1). - Does not cover systems that do not recommend an optimal path of disposition based on the cosmetic grade (Claim 3). ## The clever bit The novelty lies in combining 2D image processing with deep learning for initial defect detection and classification, then precisely measuring the 3D depth of those specific defects using an "optical spot sensor" guided by the 2D analysis. This allows for a comprehensive and automated cosmetic grade. ## Real-world examples 1. Automated inspection of smartphone displays for scratches 2. Quality control for painted car body panels 3. Sorting of pharmaceutical pills for surface imperfections 4. Grading of fruit and vegetables for blemishes 5. Inspection of semiconductor wafers for defects ## Why it matters This technology is important for quality control in manufacturing, especially for products where even small cosmetic flaws can impact value or customer satisfaction. It allows for consistent, objective, and high-speed inspection, reducing human error and labor costs. Industries producing electronics, automotive parts, or consumer goods can benefit from such automation to maintain high product standards. ## Frequently asked questions ### What does How a Robot System Inspects and Grades Objects for Flaws cover? This patent describes an automated system that uses cameras, special lighting, and deep learning AI to find and measure tiny flaws on an object's surface, then assigns a quality grade and sorts it. ### Who owns patent US 10753882? Griffyn Robotech Pvt owns this patent, granted in 2020. ### When does this patent expire? This patent is expected to expire on June 19, 2039, when the invention enters the public domain. ### What is patent US 10753882 cited by? This patent has been cited by 25 later patents that build on its ideas. ### What problem does this patent solve? This technology is important for quality control in manufacturing, especially for products where even small cosmetic flaws can impact value or customer satisfaction. It allows for consistent, objective, and high-speed inspection, reducing human error and labor costs. Industries producing electronics, automotive parts, or consumer goods can benefit from such automation to maintain high product standards. ### What does this patent NOT cover? Does not cover systems that rely solely on human visual inspection for grading objects. **Full plain-English explainer:** https://patentbrief.org/patent/us/10753882/inspection-and-cosmetic-grading-through-image-processing-system-and-method **Original patent:** https://patents.google.com/patent/US10753882 --- _Source: PatentBrief — https://patentbrief.org. Patent facts are from public records; the plain-English explanation is PatentBrief's._ ## Related patents Semantically similar inventions in the PatentBrief corpus: - [How AI Cameras Check Factory Parts for Flaws](https://patentbrief.org/patent/us/12243216/system-and-method-for-ai-visual-inspection) — This patent describes an automated system using a camera and artificial intelligence to inspect manufactured parts, identifying defect types, locations, and confidence levels, then communicating this data to a factory controller. - [How Machines Automatically Spot Flaws in Fabric During Manufacturing](https://patentbrief.org/patent/us/10942133/method-and-system-for-automatically-detecting-fabric-defect-and-computer-readabl) — This patent describes an automated system for finding defects in fabric by cleaning it, taking pictures based on how a roller spins, and then using a computer to identify flaws. - [How to Inspect Moving Objects on a Production Line with a Moving Camera](https://patentbrief.org/patent/us/20250377309/method-for-material-testing-of-an-object-in-a-production-andor-conveyor-line-and) — This patent describes a method and apparatus for inspecting specific defect locations on objects moving along a production line by having an optical camera move alongside the object to take images without stopping the line. - [How Computers Track the Position and Angle of Multiple Objects](https://patentbrief.org/patent/us/12211234/rapid-iterative-development) — A method for using multiple cameras to track the exact location and orientation of many similar objects in a 3D space, even when they overlap. - [AI That Finds Partially Hidden Objects in Images](https://patentbrief.org/patent/us/12738026/systems-methods-and-computer-programs-for-using-a-network-of-machine-learning) — This patent describes an AI system that uses multiple machine learning models to accurately identify objects in an image, even when parts of those objects are blocked by something else.