Guide d'utilisation

Getting started with CropCare AI

CropCare AI is a comprehensive agricultural technology platform that combines AI-powered plant disease detection, subscription-based premium features, and multi-platform access through mobile app and web platform. This guide will help you use all features effectively and safely.

1. Installation and setup

1.1 System requirements

• iOS 13.0+ or Android 5.0+ (API level 21+)
• Camera access permission for plant scanning
• Internet connection required for AI analysis (online service)
• Minimum 200MB free storage space for app and analysis history
• Universal Link support: https://trycropcareai.com/ deep linking
• Google Play Services for subscription management (Android)
• App Store connectivity for subscription management (iOS)

1.2 First-time setup

1. Download and install CropCare AI from Google Play Store or Apple App Store
2. Create your account using email or OAuth2 authentication
3. Grant required permissions (camera, storage) when prompted
4. Choose your region for appropriate pricing (developed/underdeveloped regions)
5. Select subscription plan: Free (5 scans/month) or Premium (unlimited)
6. Complete profile setup with your location for better recommendations

1.3 Permissions explained

• Camera: Required to capture photos of plants for AI analysis
• Storage: Needed to save analysis history and images locally using Hive database
• Internet: Essential for AI analysis via GitHub Models API and backend validation
• Location (optional): For regional pricing and environmental context in analysis
• Notifications: For subscription updates and scan limit notifications

1.4 Subscription system

CropCare uses a freemium model with regional pricing:

Free tier

• 5 AI scans per month
• Basic plant disease detection
• Treatment recommendations
• Limited history storage

FREE TRIAL (Mobile-Only)

• 30-day trial period
• 1000 AI scans per month during trial
• Full premium features including disease outbreak alerts
• Available once per user account
• No credit card required
• Activated through mobile app only
• Automatic expiry notifications at 6, 4, and 2 days before trial ends
• Seamless conversion to paid plan available anytime

PREMIUM SUBSCRIPTION (Regional Pricing)

• Unlimited AI scans
• Advanced analysis with confidence scores
• Complete treatment plans (natural & chemical options)
• Real-time disease outbreak alerts within 50km radius
• Push notifications for nearby disease outbreaks (automated daily detection)
• Disease outbreak map view showing active outbreaks
• Full analysis history with export capabilities
• Priority customer support
• Access to web platform dashboard
• Organizational features (team collaboration with license management)

Regional pricing

• Developed regions: $30.00/month, $300.00/year
• Underdeveloped regions: $5.00/month, $50.00/year
• Pricing automatically adjusted based on user region

2. Taking plant photos for best results

2.1 Photo quality guidelines

• Use good lighting (natural daylight preferred)
• Keep plant in focus with clear details visible
• Include affected areas (leaves, stems, flowers)
• Capture multiple angles if disease is present
• Use rear camera for better image quality

2.2 What to avoid

✗ Blurry or out-of-focus images ✗ Poor lighting (too dark or overexposed) ✗ Too much background clutter ✗ Extreme close-ups that miss plant context ✗ Images with fingers or objects blocking the plant

2.3 Optimal photo conditions

• Distance: 12-24 inches from plant
• Lighting: Even, natural light (avoid direct flash)
• Background: Clean, contrasting background preferred
• Angle: Include leaves, stems, and affected areas
• Time: Morning or afternoon for best natural lighting

3. Analyzing plants

3.1 Starting an analysis

1. Open CropCare AI app and ensure internet connectivity
2. Check your scan limit (shown on dashboard): Free users have 5/month, trial users have 1000/month
3. If you're a new user, you may see a free trial offer - 30 days with full premium features!
4. Tap "Take Photo" for camera or "Choose from Gallery" for existing images
5. Review the selected image and ensure good quality
6. After confirming selection, it will automatically start AI processing
7. Wait for analysis results (typically 15-45 seconds depending on connection)
8. Results are automatically saved to local history for future reference

3.2 Understanding results

The AI analysis provides comprehensive information:

• Plant Identification: Common and scientific names with confidence scores
• Disease Detection: Health status, specific disease identification, or "healthy" status
• Confidence Score: AI model's confidence level (0-100%) for reliability assessment
• Severity Assessment: Disease progression level (mild, moderate, severe)
• Detailed Symptoms: Visible signs and symptoms observed in the image
• Treatment Plans: Categorized recommendations:
• Natural/Organic treatments (eco-friendly options)
• Chemical treatments (faster-acting synthetic solutions)
• Cultural practices (environmental management)
• Prevention Tips: Proactive measures to prevent future problems
• Estimated Recovery: Timeline for treatment effectiveness

3.3 Confidence scores guide

• 90-100%: Very High confidence (reliable identification)
• 70-89%: High confidence (good identification)
• 50-69%: Moderate confidence (verify with expert)
• Below 50%: Low confidence (seek professional advice)

4. Subscription management

4.1 Accessing subscription settings

• Open app settings or profile section
• Tap "Subscription" or "Premium" to view current plan
• View remaining scans for free users
• See trial status and days remaining for trial users
• Manage subscription through Google Play Store or Apple App Store
• Access billing history and usage statistics

4.2 Activating free trial

1. New eligible users will see a trial offer dialog on the home screen
2. Trial offers full premium features for 30 days
3. Benefits include:

• 1000 AI scans per month (instead of 5)
• Disease outbreak alerts and notifications
• Advanced analytics and reporting
• Priority customer support
• Unlimited access to all premium features

1. Tap "Start Free Trial" to activate (mobile app only)
2. No credit card required for trial activation
3. You'll receive notifications at 6, 4, and 2 days before trial expires
4. Trial is available once per user account (lifetime limit)
5. Can upgrade to paid subscription anytime during trial

4.3 Upgrading to premium

1. Tap "Upgrade to Premium" from settings or when trial expires
2. Choose between Monthly or Yearly plans (save 17% with yearly)
3. Regional pricing automatically applied based on your location
4. Complete purchase through your device's app store
5. Premium features activate immediately after successful payment
6. Subscription auto-renews unless cancelled
7. If upgrading during trial, trial is marked as converted and premium access continues seamlessly

4.4 Subscription validation

• All subscriptions validated with backend server
• Purchases verified with Google Play Store or Apple App Store
• Free trial status tracked to enforce one-trial-per-user limit
• Account linking prevents unauthorized subscription sharing
• Real-time subscription status updates
• Grace period provided for expired subscriptions with temporary issues
• Trial converts automatically to free tier after 30 days if not upgraded

4.5 Managing billing

• Cancel anytime through Google Play Store or Apple App Store settings
• Refunds handled according to store policies
• Subscription remains active until end of current billing period
• Downgrade to free tier with 5 scans/month after cancellation
• Trial users revert to free tier (5 scans/month) after expiry
• No charges for trial activation - credit card not required

5. Platform access

5.1 MOBILE APP (Primary Platform)

• AI-powered scanning and analysis
• Offline result viewing (analysis requires internet)
• Local storage using Hive database and Flutter Secure Storage
• Push notifications for subscription and scan limit updates
• Free trial activation (30 days, mobile-only)
• Disease outbreak alerts and notifications (premium/trial feature)
• Real-time notifications at 6, 4, and 2 days before trial expiry

5.2 Web platform access

• URL: https://trycropcareai.com
• Advanced dashboard and analytics
• Team collaboration features for organizations
• Bulk analysis management
• API access for enterprise customers
• Comprehensive reporting and export tools
• Organisation subscription management with Paystack
• License assignment and team member access control

5.3 Notification center (premium/trial feature)

• Access disease outbreak alerts in notification tab
• Filter alerts by: All, Unread, Critical, Nearby, Read
• Alert cards show: disease name, distance, affected farmers count, severity
• Expandable details: prevention tips, treatment recommendations
• Actions: View Details, Show on Map, Mark as Read
• Swipe gestures: Right to mark read, Left to delete
• Auto-cleanup: Alerts older than 30 days removed automatically
• Local storage: Alerts saved in Hive database for offline access
• Push notifications: Real-time alerts via Firebase Cloud Messaging

5.4 Disease outbreak map view (premium/trial feature)

• Interactive map showing active disease outbreaks
• Outbreak markers colored by severity: Critical (red), High (orange), Medium (yellow), Low (green)
• Circle radius shows outbreak range (25km clustering radius)
• User location marker (blue pin) with distance calculations
• Filter by: Max distance (10-200km), Severity levels
• Tap outbreak marker to view: Disease name, case count, distance from you
• Outbreak list below map sorted by proximity
• Automatic daily updates at 2:00 AM UTC
• Requires location permission for accurate distance calculations

5.5 Deep linking and universal links

• Universal Link: https://trycropcareai.com/app/
• Custom scheme: cropcare://
• Seamless transitions between web and mobile platforms
• Shared analysis links for team collaboration
• Password reset and account verification via links

6. Interpreting treatment recommendations

6.1 Natural treatments 🌿

• Organic and eco-friendly solutions
• Safer for environment and beneficial insects
• May take longer to show results (7-14 days typically)
• Examples: neem oil, baking soda solutions, beneficial bacteria, companion planting
• Cost-effective and sustainable long-term solutions

6.2 Chemical treatments 🧪

• Faster-acting synthetic solutions (2-7 days typical response)
• More potent but may have environmental impact
• Higher cost but immediate effectiveness
• Examples: fungicides, bactericides, systemic treatments, targeted pesticides
• Professional-grade recommendations with application rates

6.3 Cultural practices 🌱

• Environmental management and prevention-focused approaches
• Irrigation management, soil amendments, crop rotation suggestions
• Long-term plant health improvement strategies
• Sustainable farming practice integration

6.4 Treatment safety

⚠️ important safety reminders:

• Always read and follow product labels and local regulations
• Wear appropriate protective equipment when applying treatments
• Consider environmental impact and effects on beneficial insects
• Test treatments on small areas first to assess plant tolerance
• Consult local agricultural extension services for region-specific advice
• Never apply treatments in windy conditions or before rain
• Respect pre-harvest intervals for food crops
• Store chemicals safely away from children and pets

7. Managing analysis history

7.1 Viewing history

• Tap "History" tab to see all past analyses stored locally
• Scroll through chronological list organized by date (newest first)
• Tap any item to view detailed analysis results and images
• Search functionality for finding specific plants or diseases
• Filter by date range, confidence level, or disease status

7.2 Data storage and sync

• All analysis data stored locally using Hive database
• Images and results cached for offline viewing
• Premium and trial users get cloud backup and cross-device sync
• Export functionality for data portability
• Analysis data includes metadata: timestamp, confidence scores, treatment effectiveness tracking
• Disease alerts stored locally with automatic 30-day cleanup

7.3 Organizing and managing data

• Create custom tags and categories for your analyses
• Mark treatments as applied and track effectiveness
• Set reminders for follow-up scans and treatment applications
• Share individual analyses with team members or experts
• Bulk export for record-keeping and reporting

7.4 Deleting data

• Swipe left on individual items to delete from local storage
• Use "Clear All History" to remove all stored data (requires confirmation)
• Deleted data cannot be recovered unless backed up to cloud (Premium feature)
• Consider exporting important analyses before deletion
• Free up storage space by removing old, resolved analyses

8. Best practices for accurate results

8.1 Photo preparation

• Clean plant leaves gently with soft cloth before photographing
• Remove dead or severely damaged material for clearer analysis
• Ensure plant is well-watered (not wilted from drought stress)
• Check multiple areas of the plant for comprehensive assessment
• Take photos during optimal lighting hours (10 AM - 3 PM)

8.2 Optimal photo techniques

• Distance: 15-30 cm from affected plant parts
• Lighting: Even, natural daylight (avoid direct flash or artificial lighting)
• Background: Clean, contrasting background (use white paper if needed)
• Focus: Ensure affected areas are sharp and clearly visible
• Angle: Capture both overall plant structure and close-up disease details
• Multiple shots: Take 2-3 angles for complex disease patterns

8.3 Environmental context

• Note recent weather conditions (rainfall, temperature, humidity)
• Document watering schedule and soil moisture conditions
• Record recent fertilizer applications or treatment history
• Consider seasonal timing and plant growth stage
• Include information about nearby plants and potential disease spread

8.4 Maximizing AI accuracy

• Upload high-resolution images (minimum 1MP recommended)
• Ensure good internet connectivity for faster processing
• Provide plant type information when prompted
• Use macro mode for detailed leaf symptoms
• Avoid heavily filtered or edited images
• Take photos in RAW or highest quality settings available

9. When to seek professional help

9.1 Critical situations requiring expert consultation

• Valuable or rare plants showing severe or unknown symptoms
• Commercial crops or food production with significant disease spread
• Rapid disease progression affecting multiple plants or crops
• Uncertainty about treatment safety, application rates, or timing
• Low confidence scores (<70%) from AI analysis
• Suspected new or emerging diseases not in AI training data
• Economic losses exceeding treatment costs

9.2 Professional resources

• Local agricultural extension services and county agents
• Certified plant pathologists and diagnosticians
• Professional horticulturists and crop consultants
• University plant diagnostic laboratories
• Agricultural cooperatives and farm advisors
• Crop insurance adjusters for commercial operations
• Integrated pest management (IPM) specialists

9.3 AI limitations to understand

• GitHub Models AI service provides preliminary assessments, not definitive diagnoses
• Cannot replace professional plant pathology expertise or laboratory testing
• Limited by image quality, lighting conditions, and camera capabilities
• May not consider complex interactions between multiple stressors
• Training data may not include all regional diseases or emerging pathogens
• Cannot assess soil conditions, water quality, or environmental factors

9.4 Geographic and regional considerations

• AI training data may have geographic bias toward certain regions
• Disease prevalence and treatment options vary by climate and location
• Local regulations may restrict certain pesticides or treatment methods
• Regional pest and disease pressure may differ from AI training data
• Always verify recommendations with local agricultural authorities
• Consider traditional knowledge and local farming practices
• Adapt treatments based on local environmental conditions

10. Troubleshooting common issues

10.1 Analysis errors and connectivity

Problem: "Failed to analyze image" or "Network connection error"

Solutions

• Verify stable internet connection (WiFi or cellular data)
• Check if you've reached scan limits (Free: 10/month, Premium: unlimited)
• Ensure image is clear, well-lit, and under 10MB in size
• Try taking a new photo with better lighting and focus
• Restart the app and attempt analysis again
• Verify subscription status if premium features expected
• Clear app cache if performance issues persist

10.2 Subscription and billing issues

Problem: "Subscription not recognized" or billing problems

Solutions

• Verify purchase completed through Google Play Store or Apple App Store
• Check subscription status in app store account settings
• For trial users, verify trial is still active and hasn't expired
• Ensure account email matches subscription purchase
• Wait 5-10 minutes for subscription validation with backend
• Try "Restore Purchases" option in app settings
• Contact app store support for refund or billing disputes
• Ensure app is updated to latest version for subscription fixes
• Remember: Each user gets only one free trial per account

10.3 Poor analysis results

Problem: Low confidence scores, unclear identification, or irrelevant results

Solutions

• Retake photo with better lighting, focus, and image composition
• Include more plant context (affected leaves, stems, overall plant)
• Ensure image shows clear disease symptoms, not just healthy parts
• Try analyzing different affected areas of the same plant
• Consider that some plant conditions may be difficult to diagnose automatically
• Consult professional help for complex or unusual symptoms
• Verify plant type selection if prompted by AI system

10.4 App performance and storage

Problem: Slow analysis, app crashes, or storage full notifications

Solutions

• Close other resource-intensive apps to free device memory
• Restart device to clear temporary files and refresh memory
• Update to latest app version through app store
• Clear analysis history to free local storage space
• Check available device storage (app requires minimum 200MB)
• Disable unnecessary background apps during analysis
• Consider upgrading device if hardware limitations persist

11. Privacy and data security

11.1 Your data and storage

• Analysis history stored locally on device using encrypted Hive database
• Images stored securely in app-specific directories with restricted access
• Authentication tokens protected via Flutter Secure Storage with hardware-backed keystore
• Personal information encrypted both at rest and in transit
• You maintain full control over data deletion and export
• No unauthorized access to your analysis data or images

11.2 Data transmission and processing

• Images transmitted securely to GitHub Models API via HTTPS encryption
• Backend validation with OAuth2 authentication
• Temporary processing only - no permanent storage of images on external servers
• Analysis results and metadata stored locally on your device
• Subscription validation handled through secure app store APIs
• Regional data processing compliance with local privacy regulations

11.3 Subscription and payment data

• Payment processing handled exclusively by Google Play Store or Apple App Store
• CropCare never stores credit card or payment information directly
• Subscription status validated through secure app store APIs
• Free trial requires no payment information - completely free for 30 days
• Trial status tracked server-side to enforce one-trial-per-user limit
• Purchase receipts encrypted and stored temporarily for validation
• Billing information managed through your app store account
• Refunds and billing disputes handled by respective app stores

11.4 Sharing and collaboration

• Analysis results can be shared through standard device sharing mechanisms
• Premium organizational features include secure team collaboration
• Data sharing requires explicit user consent and action
• Shared analysis links expire after specified timeframes
• Team data access governed by organization-level privacy settings
• Export functionality maintains data ownership and portability rights

12. Legal and safety disclaimers

12.1 Medical and safety warnings

⚠️ critical warnings:

• NEVER use CropCare AI analysis for human or animal consumption decisions
• DO NOT eat plants identified by the app without expert verification and proper identification
• Chemical treatments can be harmful to humans, animals, and environment - follow all safety guidelines
• Children and pets must be kept away from treated plants and chemical storage areas
• Always wear appropriate personal protective equipment when handling treatments
• Consult poison control immediately if accidental ingestion or exposure occurs

12.2 Accuracy and liability limitations

• AI analysis provides preliminary guidance only, not professional agricultural diagnosis
• GitHub Models API and CropCare analysis are tools to assist, not replace, expert judgment
• Confidence scores are statistical estimates, not guaranteed accuracy percentages
• Always verify important agricultural decisions with qualified plant pathologists
• Environmental factors, plant genetics, and local conditions may affect accuracy
• CropCare is not liable for crop losses, treatment failures, or economic damages
• Maximum liability limited to subscription fees paid, if any

12.3 Subscription and service terms

• Subscription plans and pricing subject to change with appropriate notice
• Service availability depends on internet connectivity and third-party API status
• Free tier scan limits enforced monthly and reset on billing cycle
• Premium features require active, validated subscription
• Service interruptions may occur due to maintenance or technical issues
• Regional pricing based on user location and may change due to economic factors

12.4 Intellectual property and data

• Users retain ownership of uploaded images and analysis data
• CropCare retains rights to improve AI models using anonymized, aggregated data
• Treatment recommendations are informational, not proprietary medical advice
• Third-party AI services (GitHub Models) subject to their terms of service
• User-generated content shared in teams governed by organization agreements

Support and feedback

For help, questions, or feedback

• In-app Help Center: Access through settings menu
• Email Support: hamez@hamez.dev (response within 24-48 hours)
• Web Platform: https://trycropcareai.com/help-center/
• GitHub Issues: For technical problems and feature requests
• API Documentation: https://trycropcareai.com/
• Community Forum: Connect with other users and share experiences

Premium Support

• Priority email support (response within 12 hours)
• Live chat support during business hours
• Direct access to agricultural consultants
• Custom feature requests and enterprise solutions

We value your feedback to continuously improve CropCare AI!

Version history and platform information

Version 1.0.1+20 (Current - October 3, 2025)

• Universal Links implementation (https://trycropcareai.com/)
• Enhanced subscription system with regional pricing
• Improved AI analysis using GitHub Models API
• Flutter Secure Storage for enhanced security
• Centralized NavigationService for better user experience
• Comprehensive subscription validation and management

Platform Details

• Minimum Android: 5.0+ (API level 21+)
• Minimum iOS: 13.0+
• Web Platform: trycropcareai.com
• AI Service: GitHub Models API
• Payment Processing: Google Play Store & Apple App Store & Paystack

Supported Crops (50+ types)

• Maize/Corn, Tomatoes, Beans, Coffee, Rice, Wheat, Potatoes
• Various vegetables, fruits, and cash crops
• Expanding database with regional crop varieties

Supported Diseases (200+ conditions)

• Fungal infections (rust, blight, powdery mildew)
• Bacterial diseases (bacterial spot, fire blight)
• Viral infections (mosaic virus, leaf curl)
• Nutrient deficiencies and environmental stress
• Pest damage and insect-related issues

[This guide will be updated as new features are added to the platform]

Retour en haut