You can deploy this model and start earning money today!

Divyanshu-Kumar19/
aapdasetu-damage-assessment

smart-india-hackathon

image-classification

mit

model-index

Views

No views yet

AapdaSetu Damage Assessment — ResNet50

Post-disaster building damage classifier for the AapdaSetu platform. Given a photo of a disaster-affected building, the model classifies damage severity into one of three grades and drives automated compensation assessment in the AapdaSetu relief pipeline.

Part of the AapdaSetu project (Smart India Hackathon).

Live Deployment

The full assessment service (this model + EXIF authenticity checks + pHash duplicate detection + compensation estimation) is deployed and publicly accessible:

🌐 https://aapdasetu-damage-api.onrender.com

Interactive demo UI: / · Swagger API docs: /docs · Health check: /health
Main endpoint: POST /api/assess-damage (multipart: photo, claimed_lat, claimed_lng, property_type, disaster_cutoff)
Production inference runs on ONNX Runtime using best.onnx from this repo (torch-free, fits Render's 512 MB free tier). Free-tier instances sleep after ~15 min idle, so the first request after a pause cold-starts in a few seconds.

Model Details


Architecture	ResNet50 (ImageNet-1K V2 pretrained backbone) + custom FC head (`Dropout → Linear(2048, 3)`)
Input	224×224 RGB image
Preprocessing	`Resize(256) → CenterCrop(224) → ToTensor → Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])`
Classes	`MINOR`, `MAJOR`, `DESTROYED` (alphabetical ImageFolder order: DESTROYED=0, MAJOR=1, MINOR=2 — see `class_mapping.json`)
Checkpoint	`best.pt` (~271 MB, PyTorch dict — keys: `model_state`, `class_to_idx`, `classes`, `epoch`, `val_acc`, `val_loss`, `optim_state`)
ONNX export	`best.onnx` (~94 MB, opset 17, softmax baked in, input `image`, output `scores`, dynamic batch) — parity with PyTorch verified (max abs diff ≈ 6e-8)

Training Recipe

Dataset: 2,400 post-disaster building images — MINOR (846), MAJOR (525), DESTROYED (1,029), split 75 / 15 / 10 per class with a leak check (image hashes verified disjoint across splits — see leak_check.json). Dataset hosted at Divyanshu-Kumar19/aapdasetu-damage-dataset.
Two-phase training:
1. Backbone frozen — only the new FC head trains
2. Full fine-tuning of the entire network
Evaluation: 5-crop Test-Time Augmentation (TTA) on the held-out test set.

Evaluation Results

Held-out test set (244 images), with TTA:

Class	Precision	Recall	F1	Support
DESTROYED	0.990	0.981	0.986	104
MAJOR	0.982	1.000	0.991	54
MINOR	0.977	0.977	0.977	86
Overall accuracy				98.36%

Full metrics: eval_report.json · Confusion matrix: confusion_matrix.png · ROC curves: roc_curves.png · Training curves: training_curves.png · Misclassified samples: misclassified.png · Training history: history.json · Split leak audit: leak_check.json

Quick Start — ONNX (recommended, CPU-friendly)

python

import json
import onnxruntime as ort
import numpy as np
from PIL import Image
from torchvision import transforms

mapping = json.load(open("class_mapping.json"))          # {"DESTROYED": 0, "MAJOR": 1, "MINOR": 2}
idx_to_class = {int(v): k for k, v in mapping.items()}

transform = transforms.Compose([
    transforms.Resize(256),
    transforms.CenterCrop(224),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
])

img = Image.open("building.jpg").convert("RGB")
x = transform(img).unsqueeze(0).numpy()

sess = ort.InferenceSession("best.onnx", providers=["CPUExecutionProvider"])
probs = sess.run(["scores"], {"image": x})[0][0]          # softmax already baked in

print("Damage grade:", idx_to_class[int(probs.argmax())])
print("All scores:", {idx_to_class[i]: round(float(p), 4) for i, p in enumerate(probs)})

Quick Start — PyTorch (`best.pt`)

python

import torch
import torch.nn as nn
from torchvision import models, transforms
from PIL import Image

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

# 1. Build the exact architecture used during training
model = models.resnet50(weights=None)
model.fc = nn.Sequential(
    nn.Dropout(p=0.0),                    # training-only; inert at inference
    nn.Linear(model.fc.in_features, 3),
)

# 2. Load the checkpoint
ckpt = torch.load("best.pt", map_location=device, weights_only=True)
model.load_state_dict(ckpt["model_state"])
model.to(device).eval()

# 3. IMPORTANT — use the checkpoint's index->class mapping.
#    ImageFolder trained classes in ALPHABETICAL order
#    (DESTROYED=0, MAJOR=1, MINOR=2), NOT ["MINOR", "MAJOR", "DESTROYED"].
idx_to_class = {int(v): k for k, v in ckpt["class_to_idx"].items()}

# 4. Preprocess exactly like training validation
transform = transforms.Compose([
    transforms.Resize(256),
    transforms.CenterCrop(224),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
])

# 5. Predict
img = Image.open("building.jpg").convert("RGB")
x = transform(img).unsqueeze(0).to(device)
with torch.no_grad():
    probs = torch.softmax(model(x), dim=1)[0]

scores = {idx_to_class[i]: round(float(p), 4) for i, p in enumerate(probs)}
print("Damage grade:", idx_to_class[int(probs.argmax())])
print("All scores:", scores)

Download the checkpoints with:

hf download Divyanshu-Kumar19/aapdasetu-damage-assessment --local-dir ./model

Intended Use

Triage of post-disaster building damage photos in the AapdaSetu claim pipeline (maps grade → compensation under NDRF/SDRF norms).
Not a substitute for a certified structural engineer's assessment.
Out of scope: images that are not disaster-affected buildings, and safety-critical decisions without human review.

Files in This Repository

File	Description
`best.pt`	Trained PyTorch checkpoint (state dict + metadata)
`best.onnx`	ONNX export for CPU inference (softmax baked in) — used by the live deployment
`class_mapping.json`	Class → index mapping (DESTROYED=0, MAJOR=1, MINOR=2)
`eval_report.json`	Per-class precision / recall / F1 on the test set
`confusion_matrix.png`	Confusion matrix visualization
`roc_curves.png`	One-vs-rest ROC curves
`training_curves.png`	Train/val loss & accuracy curves
`confidence_distribution.png`	Prediction confidence distribution
`misclassified.png`	Test samples the model got wrong
`history.json`	Per-epoch training history
`leak_check.json`	Train/val/test split leak audit

License

MIT

Divyanshu-Kumar19/aapdasetu-damage-assessment