About the project
Blind and low-vision people still need someone to tell them what is in front of them, read a medicine strip or check which note they are holding. This AI vision assistant for visually impaired Android project (DrishtiSaathi) solves that with a fully voice-controlled app. The user says "what is around me", "read this" or "which note", the camera takes a photo, Gemini Vision understands it, and the answer is spoken in Hindi, English or Marathi. For safety, one long press on SOS sends the GPS location to the family.
How it works
- The user picks "I need help seeing" and logs in with a mobile number (the number can also be spoken)
- A tap on the mic and a spoken command opens the right feature; voice words are managed by the admin
- The camera photo goes to Gemini Vision and the answer is spoken with Text-to-Speech
- Without internet, text is read on the phone (ML Kit) and notes are identified by colour
- Long press SOS → 5-second countdown → GPS location → family alert in the app + SMS with a map link
- Family members see live location, battery and SOS history on a map and mark alerts resolved
- The admin manages users, SOS alerts, voice commands, note colours, Gemini settings and reports
Users
- Visually impaired user: voice commands, scene description, text reading, currency identification, SOS, emergency contacts, history
- Family member: link with a user code, live location map, SOS alerts and notifications
- Admin (web): dashboard, users, family members, SOS alerts, usage logs, voice commands, currency notes, reports, AI settings
Highlights
- 3 roles, 18 Android screens, 11 admin pages, 23 PHP APIs, 12 database tables
- Accessibility-first UI: big coloured rows, spoken labels, large mic and SOS buttons
- Gemini Vision with an offline fallback (ML Kit OCR + colour check for notes)
- Validations on the app and server, bcrypt passwords, prepared statements
- Sample data and demo logins included, ready for the demo
What you get
- Complete Android Studio source code (Java) and PHP admin panel + APIs
- MySQL database file with sample data
- Project report (Word, 50+ pages) with DFD, ER, UML and deployment diagrams
- 2-page project abstract, project explanation with viva questions
- Step-by-step setup guide and setup help
Who can use this project
Suitable as a final year project for BE / B.Tech (Computer, IT, ENTC), MCA, BCA and Diploma students. A strong choice if you are looking for an AI Android project with source code, an assistive technology project for blind people, or a Gemini AI app project with documentation and report.
Frequently asked questions
Does the app work without internet? Yes, partly. Text reading and currency identification have offline modes; the full scene description needs internet.
Is the Gemini API paid? No. A free Gemini API key from Google AI Studio is enough for the demo; the admin pastes it in Settings.
Can the family member get alerts on their own phone? Yes. The family member links with the user's code and gets SOS alerts with the location in the same app.
Which languages are supported? Answers can be spoken in Hindi, English or Marathi, and the speech speed can be changed.
