Market Application Background
To ensure the reliable operation of the optimizer series products, comprehensive and rapid inspection of their complex structure and minute features is necessary. AI-powered visual appearance inspection solutions can efficiently cover multiple types of defective appearances, accurately identify subtle defects, effectively improve production line first-pass yield, and reduce quality risks.
▲ Optimizer Material Diagram ▲
Positive motion AI visual appearance inspection solution
The AI-powered visual appearance inspection solution integrates high-precision image recognition and automation technology, and is integrated into the MG series fully automatic packaging inspection machine. By processing images through appearance inspection software and utilizing multi-dimensional inspection tools such as grayscale values, area, similarity, and edge detection (calipers), it accurately determines the compliance of product appearance. The newly upgraded optimized packaging inspection machine AOI system provides customers with efficient, stable, and zero-missing-detection inspection assurance.
▲ Packet inspection ▲
Advantages of AI Visual Appearance Inspection Solution
● Millisecond-level detection and massive storage: Single-sided image processing speed <0.1 seconds, supports real-time monitoring of 4 screens, and automatically stores more than 50,000 defect maps.
● High efficiency and stability, comprehensive coverage: single-piece inspection cycle time ≤ 4.5 seconds, equipment utilization rate ≥ 90%, meeting the needs of high-load production lines; multi-sided synchronous inspection (bottom/tail adhesive/pin/printed surface), completing full appearance coverage of the product in one go.
● Refined management and convenient operation: Provides three-level user permission control (operator/technician/engineer), and the Chinese interface supports flexible parameter adjustment.
01. Application of AI visual appearance inspection in fully automated package inspection machines
Design of a Fully Automated Packaging Inspection Solution Based on Positive Motion Technology
System Components
The AI visual appearance inspection system consists of five core modules that work together to form a complete intelligent inspection closed loop:
1. Image Acquisition Module: Integrates industrial-grade imaging components such as industrial cameras, lenses, and light sources to accurately capture image data of product surfaces;
2. AI Image Processing Platform: Based on deep learning algorithms, it performs real-time processing and analysis of acquired images, efficiently performing defect identification (such as scratches, cracks, stains, and holes) and high-precision size/shape measurement tasks;
3. Model Training Module: Provides end-to-end tools from data annotation and model training to deployment and validation, supporting users to independently build customized detection models;
4. Result Output Module: Presents real-time visual inspection results and can interact with the production control system;
5. System Operation and Maintenance Management Module: Monitors the system's operating status and provides functions such as parameter configuration and fault diagnosis.
AI visual inspection process implementation
The system is equipped with a high-resolution industrial camera and a dedicated light source, enabling it to capture high-definition images of passive components. Upon receiving a trigger signal, the system activates the light source and acquires images via the industrial camera. The acquired images are then analyzed and inspected, and the inspection results are output.
The system's interface is simple and intuitive. Operators can modify parameters, trigger detections, and view results using a touchscreen or mouse and keyboard.
▲ Detection interface diagram ▲
Test case illustration
The inspection items include core misalignment, core damage, base misalignment, cracked tail adhesive, insufficient tail adhesive, solder buildup on leads, poor lead coplanarity, lead misalignment, and lead blackening, etc. The inspection illustrations are shown below:
Core functions of AI visual appearance inspection system
1. Precise Defect Identification: Comprehensive detection of various defects on product surfaces, such as scratches, cracks, stains, and holes, and accurate recording and analysis of key information such as defect type, location, and size, effectively preventing defective products from leaving the product;
2. Dimension Measurement: Automatically measures key dimensions of the product such as length, width, height, and diameter, compares them with standard data in real time, and quickly determines the product's dimensional compliance;
3. Shape Consistency Judgment: Precisely inspect the product's geometric features (such as roundness, straightness, and flatness) and compare them with standard shapes to ensure that the product shape meets the requirements;
4. Color Quality Control: Accurately detect color differences and color deviations in products, compare with standard color sample data, and ensure batch-to-batch color consistency and quality.
5. Intelligent data management: Real-time storage, query, statistics and analysis; production inspection reports and quality analysis reports provide strong data support for production process monitoring and quality control.
Solution Operation Process
Advantages of the solution
1. Ultra-high precision and low false positive rate: By integrating deep learning and traditional vision algorithms, it achieves a false positive rate of ≤1% and a false negative rate of ≤0.1% (with a fatal defect false negative rate of 0%), accurately identifying a variety of minute defects;
2. Increased Production Capacity: Supports parallel processing of multiple tasks, resulting in fast testing speeds and significantly improved production line efficiency;
3. Zero programming required and easy to operate: No programming knowledge is needed; the user interface is intuitive and user-friendly, significantly reducing equipment debugging and personnel training time.
4. Deeply customized services: We flexibly customize the software interface and core functions according to the specific needs of our clients to meet diverse testing scenarios;
5. Intelligent Data Empowerment: Provides powerful statistical analysis and reporting functions, seamlessly integrates with MES systems, and drives production process optimization;
6. Excellent compatibility: Supports multiple mainstream communication protocols and hardware devices, easily adapting to different brands of testing equipment.
Solution Hardware Configuration
02. AI Visual Appearance Inspection System
Zhengdong Technology's AI-powered visual appearance inspection system, based on machine (deep) learning technology, enables automated and high-precision inspection of product appearance. This system can efficiently identify various surface defects (such as scratches, cracks, stains, and holes) and accurately measure key appearance features such as product dimensions and shape. It boasts significant advantages such as fast inspection speed, high accuracy, and good stability, and is widely used in quality control processes across numerous industries including electronics, automotive, machinery, food, and pharmaceuticals.
Technical support
● Stringent acceptance standards: Before system delivery, it must pass multiple tests, including mixed package verification, load testing, and no less than one month of commercial trial operation, to ensure equipment stability and a test pass rate of ≥99.8%.
● Full-service support: We provide one-stop services such as installation, commissioning, and technical training to help customers seamlessly connect and quickly put their products into production.
To help you better and faster develop and apply your AI visual appearance inspection solution, we have prepared detailed documentation. For further information on usage, parameter configuration, and practical application examples, please contact us at 400-089-8936.
