-
AI User Manual
FAQ
General Questions
Q1: What is AeroMegh Intelligence ?
📊 Answer: AeroMegh Intelligence is an AI-powered platform for managing projects, detectors, and reports. It helps users analyze data, train AI models, and generate insights efficiently.
Dashboard
Q2: What does the Dashboard display?
📊 Answer: The Dashboard provides an overview of key metrics such as:
- Number of images uploaded.
- Number of detectors created.
- Storage usage.
- Global detection map for geographical insights.
Q3: How do I return to the Dashboard from other sections?
🔝 Answer: Click the “Dashboard” tab in the top navigation bar.
Projects
Q4: How do I create a new project?
➕ Answer:
- Tap the “+” button in the Projects section.
- Enter a project name (e.g., Farm Analysis 2024).
- Upload images and define the project area.
- Click “Save” to finalize.
Q5: How do I locate an existing project?
🔍 Answer: Use the Search Bar in the Projects section to type the project name or keyword.
Q6: How do I open a project?
📂 Answer:
- From the Dashboard: Click the “Open” button next to the project name in the Recent Projects panel.
- From the Projects List: Select the project name to open it.
Q7: What information is displayed for each project?
📋 Answer: Each project displays:
- Name
- Creation date
- Area covered
- Number of linked images
Detectors
Q8: What is a Detector?
📊 Answer: A Detector is an AI model trained to identify specific objects or patterns in images (e.g., detecting “cauliflower” in agricultural datasets).
Q9: How do I create a new Detector?
🛠️ Answer:
- Tap “+” in the Detectors section.
- Assign a name and define the object classes (e.g., Vegetables).
- Upload training data.
- Define and annotate areas (e.g., Training, Testing, Accuracy).
- Click “Train Detector” to finalize.
Q10: What do the status labels on detectors mean?
✅ Completed: Detector is trained and ready for use.
⚠️ Requires Training: Detector needs more data or annotations.
Q11: How do I assign a Detector to a project?
🎯 Answer:
- Open the project.
- Click “Assign Detectors” in the Detectors section.
- Select a detector and click “Save.”
Q12: How can I improve Detector accuracy?
🎯 Answer:
- Use diverse and high-quality images for training.
- Regularly retrain detectors with updated data.
- Remove irrelevant or low-quality data.
Reports
Q13: What is the purpose of the Reports Screen?
📊 Answer: It lists all generated reports with options to view, download, or refresh their status.
Q14: How do I search for a specific report?
🔍 Answer: Use the Search Bar to enter keywords like report name, detector name, or object class.
Q15: What does the Monitoring Report Screen display?
📖 Answer:
- Report name
- Detected object classes
- Visuals or 3D models
- Pagination for multiple detection pages
Q16: How do I generate and download a report?
📑 Answer:
- Click “Generate Report” to create it (if incomplete).
- Once completed, click “Download” to save the report as a PDF.
Workflow and Usage Tips
Q17: What is the recommended workflow for using AeroMegh Intelligence ?
🛠️ Answer:
- Create a Project: Add images and define project areas.
- Assign Detectors: Select AI models for analysis.
- Mark Detection Areas: Use the Pen Tool to highlight zones on images.
- Run Analysis: Use the “Detect on Entire Project” button.
- Review Results: Check data in the Results tab or Reports Screen.
Q18: How do I mark detection areas in an image?
✏️ Answer:
- Open the image or click “View on Map.”
- Use the Detection Area Tool to draw detection areas.
- Save and proceed with the detection task.
Q19: How do I track the progress of a detection task?
🔄 Answer: Open the Results Section in the Project View to check the status (In Progress or Completed).
2. Select Detectors
- In the Assign available detector Screen, detectors are displayed as cards containing the following details:
- Detector Name: The name of the detector (e.g., Pit2, Pits_Detector).
- Accuracy: Indicates the performance accuracy of the detector. If unavailable, it is marked as NA.
- Classes: Displays the types of data the detector can process (e.g., “Pit,” “Cauliflower”).
- Status:
- Completed: Ready for use.
- Requires Training: The detector needs training before usage.
- Date: Displays the creation or training date of the detector.
Steps to Select a Detector
- Mark the Checkbox: Click the checkbox in the top-right corner of the desired detector card.
- Multi-Selection: Select multiple detectors by marking additional checkboxes.
- Direct Assignment:
- If a detector has already been used for the image and is available under the Detectors tab, you can directly check and assign it on the image’s screen by selecting it from the list.
- Assign Detectors
- After selecting detectors, click the Assign button at the bottom of the screen to finalize the selection.
Additional Options
- These detectors are ready to be applied to images. Go to Detectors or Click here if you want to train a new one.
- Clicking on Detectors will navigate to the Detectors page, where you can explore and manage all available detectors.
- Clicking on Click here will open the Train a New Detector dialog.
- To train a new detector manually, visit the Train New Detector section for detailed instructions.
🎯 Tips for Effective Detector Selection:
- Accuracy: Prefer detectors with higher percentages for more precise analysis.
- Classes: Ensure the detector supports the required data type.
- Status: Use only detectors labeled as “Completed” for immediate results.
- Navigation:
- Use the pagination feature at the bottom to browse through multiple detector pages.
- The “+” button allows you to create or add a new detector if needed.
Additional Tips and Features
Q20: What do the icons mean on the platform?
- 🟢 Green: Indicates readiness or completion.
- ⚠️ Yellow: Requires action or input.
- 📊 Graph: Represents performance or accuracy metrics.
- 📂 Folder: Denotes image or project data.
Q21: Can I export projects or data?
📤 Answer: Yes, you can export reports as PDF files. Plans for exporting projects as CSV files may be available in future updates.
For further assistance, reach out to support@pdrl.in. Let us know if additional details or troubleshooting steps are needed! 🚀