Adicionar arquivos iniciais

Dockerfile

@@ -0,0 +1,18 @@
+FROM python:3.13.3
+
+RUN useradd -m -u 1000 user
+RUN apt-get update && apt-get install -y \
+    python3 \
+    python3-pip \
+    libgl1
+USER user
+ENV PATH="/home/user/.local/bin:$PATH"
+
+COPY . /app
+
+WORKDIR /app
+
+RUN pip install --no-cache-dir --upgrade -r requirements.txt
+
+COPY --chown=user . /app
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

app.py

@@ -0,0 +1,36 @@
+# uvicorn app:app --reload
+
+from fastapi import Body, FastAPI, UploadFile, File
+from fastapi.responses import JSONResponse
+from fastapi.middleware.cors import CORSMiddleware
+from detect import DengueDetector
+import traceback
+
+# Inicializar detector e preditor
+detector = DengueDetector()
+app = FastAPI()
+
+# --- CORS ---
+origins = ["https://previdengue.vercel.app", "*"]
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=origins,
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"]
+)
+
+# --- Rotas ---
+@app.get("/")
+def health_check():
+    return {"status": "ok", "message": "API de Dengue rodando!"}
+
+# --- Rota de detecção ---
+@app.post("/detect/")
+async def detect(file: UploadFile = File(...)):
+    try:
+        content = await file.read()
+        result = detector.detect_image(content)  
+        return JSONResponse(content=result)
+    except Exception as e:
+        return JSONResponse(status_code=500, content={"error": str(e)})

detect.py

@@ -0,0 +1,54 @@
+from collections import Counter
+import numpy as np
+from PIL import Image
+from io import BytesIO
+from ultralytics import YOLO
+
+class DengueDetector:
+    def __init__(self, model_path="./models/DetectsmallTest1.pt"):
+        self.model = YOLO(model_path)
+        self.names = self.model.names
+
+    def calculate_intensity(self, objects):
+        weights = {"piscina": 9, "caixa_agua": 4, "carro": 1}
+        score = sum(weights.get(obj["class"], 0) for obj in objects)
+        return score
+
+    def detect_image(self, image_bytes):
+        # Carregar imagem da memória
+        img = Image.open(BytesIO(image_bytes)).convert("RGB")
+        img_np = np.array(img)  # YOLO aceita np.array diretamente
+        height, width = img_np.shape[:2]
+
+        # Detectar objetos
+        results = self.model(img_np)
+        result = results[0]
+        boxes = result.boxes
+        class_ids = boxes.cls.tolist()
+        confidences = boxes.conf.tolist()
+        class_names = [self.names[int(cls)] for cls in class_ids]
+        counts = Counter(class_names)
+
+        # Construir lista de detecções
+        detections = []
+        for i in range(len(boxes)):
+            x1, y1, x2, y2 = map(float, boxes.xyxy[i])
+            conf = float(confidences[i])
+            cls_id = int(class_ids[i])
+            detections.append({
+                "class": self.names[cls_id],
+                "confidence": round(conf, 4),
+                "box": {
+                    "x1": x1, "y1": y1, "x2": x2, "y2": y2,
+                    "original_width": width, "original_height": height
+                }
+            })
+
+        intensity_score = self.calculate_intensity(detections)
+
+        return {
+            "total": len(class_ids),
+            "contagem": counts,
+            "objetos": detections,
+            "intensity_score": intensity_score
+        }

requirements.txt

@@ -0,0 +1,10 @@
+fastapi==0.115.12
+uvicorn==0.34.0
+pillow==11.1.0
+opencv-python-headless
+numpy==2.1.1
+ultralytics==8.3.105
+torch==2.6.0
+python-multipart
+pandas==2.2.3
+pyarrow==20.0.0