Galileo series AI vision software solution

Galileo series software


Based on the application scenarios of fast deep learning technology in industrial vision and industrial defect detection, we launched "Galileo" series of software products.This series of products will be "AI deep learning algorithm, model training, model testing, online detection and analysis, data analysis report, automatic generation of defects, cloud deep learning service, AI customized algorithm" and other one-stop implementation.Different functional modules and different products can be selected according to the different needs of customers' actual use scenarios

Deep learning system

Cloud platform for industrial visual

Online detection and big data system

Defect sample management system

Software testing process
Introduction of software

Deep learning system:Galileo-X

Core functions: modeling, annotation, training, verification, detection, feedback, additional training, reporting

GalileoX independent research and development in both China and the United States is a deep innovation team, fully independent intellectual property rights, based on the technology of "rapid deep learning" industrial visual identification system, using the deep learning technology in industrial visual identification detection, classification, location, OCR scene demand one-stop complete solution, and greatly reduce maintenance difficulty and cost in the late algorithm。




DetectionClassificationLocationOCR

Online detection system:Galileo-T

Core functions: data statistics, back check, analysis, reporting, decision support/marking overkill, omission/feedback training

Online detection:The product is designed for the real-time operation and detection of the production line. It serves the front end of the industrial testing equipment and directly faces the equipment operator, providing great convenience for the production line detection work。

Function and extensibility:The integration of detection module, defect re-filtering module and data sample feedback module enables perfect docking with GALILEO-X or similar platforms, as well as all operation and control systems, industrial control systems and equipment on the market。

Free custom:According to products to independent standards and forms are more flexible。

Production line testingOriginLab OriginProFreedom to buildInfinite compatible

Big data aided decision making system:Galileo - T

GalileoT Big data aided decision making system is mainly used in the big data analysis of industrial product defect detection, which can more comprehensive and efficient statistics and management of production and testing data 。

Product testing data plays a crucial role in the quality of product production and shipment. In a large number of fragmented data, it is difficult to locate the connection status of each link. Therefore, the big data decision system was born. It is designed to solve the problem of isolated data information of production line, summarize and analyze the test data of all production lines and all links of the factory, and give real-time alarm to abnormal data to improve product quality and reduce unnecessary losses 。


Big Data AnalysisAid Decision MakingTotal Data SummaryReal-time exception warning

大数据算法无缝匹配
多种样本生成方案
多平台无缝对接

Defect sample management system:Galileo-D

Core functions: Defect library classification, defect generation, new sample generation, data invocation, training, testing

Specially designed for auxiliary sample collection and defect labeling, it reduces more than 80% manual operation and 90% sample collection time, thus realizing the "no sample training model" in a real sense. 。






Huge count of sampleSeamless matchFree call

Product Advantage
  • Productiz covenant-lite

    Designed for local production enterprises, very simple production line workers maintenance interface

  • High efficiency

    Fast modeling and labeling, saving 80% of the time;Customization greatly improves the detection speed

  • High speed neural network

    Self-research core technology,optimized neural network,model accelerates over 30~50 times

  • High accuracy

    Through addition and feedback training, the detection accuracy is close to 100%

  • Whole process inspection

    Constructing four whole process testing system, the industrial application is more extensive

  • Support for complex images

    Sensitive detection under complexity background, beyond the existing traditional image analysis, detection technology

  • Cross-platform Compatibility

    The new framework of deep learning can be docked and transplanted to any software and hardware platform

  • Visual report

    Support the output of custom, visual test and detection reports

  • Over 2 billion sample bases

    More than 2 billion samples of industrial production defects

  • Setting standards flexibly

    Adjust positive and negative samples according to production requirements, conduct optimization training

  • Online quality inspection

    Multi-equipment on-line quality inspection big data analysis, to provide reliable decision aid for industrial production

  • MES System

    Docking MES system

Some application cases

2.5D / 3D Phone Cover Glass

Surface detection

Speed:≤ 2.3s/pc

Result:Loss ≤ 2.5% | Fault ≤ 10%

Type:Smudge, hair, dust spot, scratch, edge collapse, bump, deep scratch......

Defect Type:deep scratch
  • master map

  • result

Metallized Ceramics

Surface detection

Speed:≤ 100ms/pc

Result:Loss ≤ 0.3% | Fault ≤ 1%

Type:Scratches, chipped edges, smudges, cracks, scratches, missing......

Defect Type:scratch
  • master map

  • result

Metallic Part

Surface detection

Speed:≤ 500ms/pc

Result:Loss ≤ 0.1% | Fault ≤ 5%

Type:Smudge, bright mark, scratch, spot, mold, knife grain, notch......

Type:Burr、scratch
  • master map

  • result

Industrial character recognition

Surface detection

Speed:≤ 120ms/pc

Result:Precision rate ≥ 99.99%

Defect Type:Capital
  • master map

  • result

Electronic Components

Surface detection

Speed:≤ 2ms/pc

Result:Loss ≤ 0.01% | Fault ≤ 1%

Type:Magnetic leakage,bubbles,cracking,deformation,black,crack,crescent,trachoma,indentation......

Defect Type:End of bad
  • master map

  • result

Pouch power cell

Surface detection

Speed:≤ 5s/pc

Result:Loss ≤ 0.1%(serious defects is 0) | Fault ≤ 5%

Type:Polar piece folding, breakage, leakage, edge sealing foreign body,convex, pinhole......

Defect Type:Sealing foreign body
  • master map

  • result