Update app.py

app.py

@@ -4,21 +4,45 @@ from ultralytics import YOLO
 import requests
 import json
 import logging
+import cv2
 
 logging.basicConfig(level=logging.INFO)
 
-model = YOLO("ART_Positive_Negative_Detection_v1.pt")
+model_detection = YOLO('/content/drive/MyDrive/Databae/ART/runs/train/yolov8_detection/weights/best.pt')
+model_classification = YOLO('/content/drive/MyDrive/Databae/ART/runs_positive/train/yolov8_classification/weights/best.pt')
 
 def detect_objects(images):
-    classes={ 0: "Positive", 1: "Negative"}
-    names = []
-    results = model(images, conf=0.3)  
-    for result in results:
-        if len(result.boxes) > 0:
-            class_index = result.boxes.cls[0].item()
-            class_name = classes[class_index]
-            names.append([class_name])
-            # print(result)
+    classes={ 2: "Positive", 1: "Negative"}
+    names = [] 
+    results_detection = model_detection(image_path)
+
+# Load the image using OpenCV
+    # img = cv2.imread(image_path)
+    print(type(img))
+    # Process each detected object
+    for result in results_detection:
+        for box in result.boxes:
+            # Get bounding box coordinates (x1, y1, x2, y2)
+            x1, y1, x2, y2 = map(int, box.xyxy[0])
+    
+            # Crop the bounding box from the image
+            cropped_img = img[y1:y2, x1:x2]
+    
+            # Resize the cropped image to 640x640
+            resized_img = cv2.resize(cropped_img, (640, 640))
+            resized_img = cv2.cvtColor(resized_img, cv2.COLOR_BGR2RGB)
+            # Perform inference using the classification model
+            results_classification = model_classification.predict(resized_img,save = True)
+    
+            detected = False
+            print(results_classification)
+            # Process classification results
+            for res in results_classification:
+                # Get class probabilities and labels
+                top1_class = res.probs.top1  # Predicted class
+                top1_confidence = res.probs.top1conf
+                print(top1_class)
+                name.append([classes[top1_class]])
         else:
             names.append(['None'])
         # print(names)