-01.png){kind=small}
Visual inspection solution that automates visual inspection using AI
AI is
Make it yourself
Towards a new era
Creating AI in the production field
Easy edge implementation and operation!
AI generation tool to solve challenges in the manufacturing industry
Factory optimization
MANUFACIA's anomaly detection and predictive maintenance capabilities help manage costs and optimize production infrastructure.
Quality improvement/stabilization
By incorporating the inspection skills of experienced personnel into the AI, we eliminate the variability in judgments caused by human error!
From dataset registration, model configuration, training, to implementation.
MANUFACIA-Vision's user-friendly UI and data visualization capabilities allow for quick and easy data quality control.
STEP 1
Selecting a task
This creates a project to generate an AI model. You can specify any project name and directory.
STEP 2
Load Your Training Data
Register the datasets used for training, validating, and inferring AI models. You can register them by dragging and dropping from a directory.
STEP 3
Pre-Process Automatically
Select the AI algorithm and feature extraction capabilities. You can also configure training options and label the dataset.
STEP 4
Learning
The AI model will be trained according to the model settings. Once training is complete, a summary of the model's performance will be reported.
STEP 5
Visualize & Select Your Model
The validation results of the trained AI model will be reported. You can change the thresholds and output the report as a PDF.
STEP 6
Implementing AI
We analyze the generated AI model and implement the optimal model into the production line based on sample code and other resources.
MANUFACIA-Vision Operation Image
Conventional visual inspection methods faced the challenge of over-detection, which resulted in many good products being mistakenly classified as defective.
Cross Compass's AI-powered visual inspection solution supports this improvement, reducing the time-consuming and laborious process of manually re-inspecting and visually checking products.
At the same time, it will also enable the transfer of the intuition and expertise of skilled inspectors through digital technology.
We work with our customers to solve on-site challenges, from the gradual implementation of AI technology to achieving cost reductions.
The process leading up to implementation
Technical Verification
Possibility assessment
AI Software Development
1 week
1 week
Confirming the effectiveness of AI in resolving inspection challenges.
We will verify the effectiveness of AI utilization and propose implementation and application plans.
System implementation and development
AI Software Development
AI Software Development
1-3 months
1 month ~
We develop user-friendly custom software tailored to your operations, from data integration and AI analysis to result display and storage.
Building an integrated environment between AI generation tools and production site systems.
Operation
Full operation
AI Software Development
1-3 months
Continuous support
We provide continuous improvement and updates to AI models, as well as technical support.
Supporting long-term stable operation, including system updates.
Visual Inspection: Application Examples and Features
①
From Evaluation to Implementation
One-stop serviceFeasibility → Imaging Setup → Accuracy Testing → AI Software → System Deployment → On-site Operation
②
AI Generation with Good Data OnlyGenerate highly accurate AI models using only good-product data, even when defect data is difficult to collect.
③
Visualize AI ResultsVisualize anomalies with heatmaps and save the results as quality data.
④
Easy AI Generation & Improvement On-SiteAnyone can quickly generate and improve AI models, enabling “AI that grows on the shop floor.”
Medical devices
Film sheet
Electronic circuit
board
Mural materials
_edited_edited.jpg)
semiconductor
material
Industrial parts
