A full-stack image analysis tool built in Go that performs real-time technical quality inspection — sharpness, exposure, noise, color profiling, and EXIF metadata extraction — through a clean, mobile-responsive web UI.
Live Demo: image-analyzer-nmri.onrender.com
Upload Analysis Result EXIF Metadata
Batch Queue Batch Results
Upload any JPEG or PNG → get back a full technical quality report with an overall score, visual metrics, and camera metadata.
| Metric | Method |
|---|---|
| Sharpness | Laplacian variance — detects blur at pixel level |
| Brightness | Average luminance — flags underexposed / overexposed |
| Noise | Neighbor-diff Laplacian — detects high-frequency random variation |
| Color Profile | Dominant hue, vibrance, avg RGB |
| EXIF Metadata | Camera model, ISO, focal length, exposure time |
| Layer | Technology |
|---|---|
| Backend | Go (net/http) |
| Image Processing | golang.org/x/image, goexif |
| EXIF (mobile) | exif-js — client-side extraction before upload |
| Frontend | Vanilla JS, CSS Grid, SVG animations |
| Deployment | Render.com |
Browser (upload)
│
├── exif-js reads EXIF from original file (before resize)
│
├── Canvas resizes image if > 1600px
│
└── POST /analyze { image, client_exif }
│
Go Server
├── ReadExif() ← goexif on raw bytes
├── DetectBlur() ← Laplacian variance
├── CheckBrightness() ← avg luminance
├── DetectNoise() ← neighbor diff
└── AnalyzeColor() ← dominant hue + vibrance
(all 5 run concurrently via goroutines)
Concurrent analysis — all 5 analyzers run in parallel goroutines using sync.WaitGroup, making analysis fast even for large images.
EXIF on mobile — Android Chrome strips EXIF metadata during file upload (privacy feature). Fixed by reading EXIF client-side with exif-js before the canvas resize destroys it, then sending it as a fallback field alongside the image.
Raw bytes preserved — imageData []byte is kept separately from the decoded image.Image, so EXIF (which lives in the raw file bytes, not pixel data) is never lost during image decoding.
Batch processing — up to 20 images processed concurrently server-side, each in its own goroutine.
Analyzes a single image.
Request: multipart/form-data
image— JPEG or PNG file (max 10MB)client_exif— JSON string with EXIF data (optional, mobile fallback)
Response:
{
"file_name": "photo.jpg",
"width": 1200, "height": 1600,
"blur": { "sharpness": 0.06, "status": "sharp" },
"brightness": { "value": 53.94, "status": "good" },
"noise": { "level": 0.03, "status": "moderate" },
"exif": {
"camera": "samsung SM-F415F",
"iso": "25",
"focal_length": "5mm",
"exposure_time": "0.02s"
},
"overall_score": 90
}Analyzes up to 20 images concurrently.
Exports analysis results as a downloadable CSV.
git clone https://github.com/anjali-rayy/image-analyzer
cd image-analyzer
go run main.goOpen http://localhost:8080 in your browser.
image-analyzer/
├── main.go # HTTP server, routing, batch handler
├── analyzers/
│ ├── blur.go # Laplacian sharpness detection
│ ├── brightness.go # Luminance analysis
│ ├── noise.go # Noise estimation
│ ├── color_profile.go # Dominant color + vibrance
│ └── exif.go # EXIF metadata extraction
├── static/
│ ├── index.html
│ ├── script.js # Upload, EXIF client-side, results UI
│ └── style.css # Dark theme, mobile-responsive
└── go.mod