Application

app.py

@@ -0,0 +1,81 @@
+import os
+from pathlib import Path
+import gradio as gr
+import cv2
+import numpy as np
+import torch
+import spaces
+from yolox.exp import get_exp
+from yolox.utils import fuse_model, postprocess, vis
+from yolox.data.data_augment import preproc
+from PIL import Image
+from pathlib import Path
+
+MODEL_PATH = "models/yolox-tiny.pth"  # Path to trained model
+EXP_FILE = "exp.py"  # Path to your experiment file
+CONF_THRESHOLD = 0.4  # Confidence threshold
+NMS_THRESHOLD = 0.65  # Non-max suppression threshold
+
+# Load experiment
+@spaces.GPU
+def process_frame(frame):
+    exp = get_exp(EXP_FILE, None)
+    model = exp.get_model()
+    model.eval()
+
+    ckpt = torch.load(Path(MODEL_PATH), map_location='cpu')
+    model.load_state_dict(ckpt["model"])
+    model = fuse_model(model)
+    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+    model.to(device)
+
+    # pil_image = PIL.Image.open('Image.jpg').convert('RGB')
+    open_cv_image = np.array(frame)
+    # Convert RGB to BGR
+    img = open_cv_image[:, :, ::-1].copy()
+
+    img_input, ratio = preproc(img, exp.test_size)
+    img_input = torch.from_numpy(img_input).unsqueeze(0).float().to(device)
+
+    with torch.no_grad():
+        outputs = model(img_input)
+        outputs = postprocess(outputs, exp.num_classes, CONF_THRESHOLD, NMS_THRESHOLD)
+
+    if outputs[0] is not None:
+        dets = outputs[0].cpu().numpy()
+        bboxes = dets[:, :4].astype(int)
+        scores = dets[:, 4]  # Achte darauf, dass der Index für Scores korrekt ist
+        cls_ids = dets[:, 5].astype(int)
+
+        result_img = vis(img, bboxes, scores, cls_ids, class_names=exp.class_names, conf=CONF_THRESHOLD)
+    else:
+        result_img = img  # No detections, return original frame
+
+    return result_img
+
+def get_default_image_paths(folder_path):
+    image_extensions = ('.jpg', '.jpeg', '.png', '.gif', '.bmp', '.tiff')
+    image_paths = [[os.path.join(folder_path, file)] for file in os.listdir(folder_path) 
+                   if file.lower().endswith(image_extensions)]
+    return image_paths
+
+default_images = get_default_image_paths(Path("examples/"))
+
+def process_input(file_input):
+    processed_img = process_frame(file_input)
+    processed_img = processed_img[:, :, ::-1].copy()
+    return Image.fromarray(processed_img)  # Return the processed image directly
+
+# Create Gradio Interface with title and description
+iface = gr.Interface(
+    fn=process_input,
+    inputs=[
+        gr.Image(label="Upload Image", type="pil"),  # File input as PIL Image
+    ],
+    outputs=gr.Image(type="pil", label="Output (Image)"),  # Show output as an image
+    examples=default_images,
+    title="Strawberry Disease Detection",
+    description="This application detects diseases in strawberries using a trained YOLOX model. Upload an image, video, or use your webcam for analysis."
+)
+
+iface.launch(share=True)

exp.py

@@ -0,0 +1,28 @@
+import os
+from yolox.exp import Exp as MyExp # type: ignore
+#from yolox.data import COCODataset   # type: ignore
+
+class Exp(MyExp): 
+    def __init__(self):
+        super(Exp, self).__init__()
+        self.exp_name = os.path.split(os.path.realpath(__file__))[1].split(".")[0]
+        self.num_classes = 7
+        self.class_names=["Angular Leafspot", "Leaf Spot", "Anthracnose Fruit Rot", "Blossom Blight", "Gray Mold", "Powdery Mildew Fruit", "Powdery Mildew Leaf"]
+
+        # small
+        # self.depth = 0.33 
+        # self.width = 0.50
+
+        # tiny
+        self.depth = 0.33
+        self.width = 0.375
+        self.input_size = (416, 416)
+        self.mosaic_scale = (0.5, 1.5)
+        self.random_size = (10, 20)
+        self.test_size = (416, 416)
+        self.enable_mixup = False
+
+        self.data_dir = "coco_dataset"
+        self.train_ann = "instances_train.json"
+        self.val_ann = "instances_val.json"
+        self.test_ann = "instances_test.json"

requirements.txt

@@ -0,0 +1,102 @@
+absl-py==2.3.1
+aiofiles==23.2.1
+annotated-types==0.7.0
+anyio==4.5.2
+cachetools==5.5.2
+certifi==2025.7.14
+charset-normalizer==3.4.2
+click==8.1.8
+contourpy==1.1.1
+cycler==0.12.1
+exceptiongroup==1.3.0
+fastapi==0.116.1
+ffmpy==0.5.0
+filelock==3.16.1
+flatbuffers==25.2.10
+fonttools==4.57.0
+fsspec==2025.3.0
+google-auth==2.40.3
+google-auth-oauthlib==1.0.0
+gradio==4.44.1
+gradio_client==1.3.0
+grpcio==1.70.0
+h11==0.16.0
+hf-xet==1.1.5
+httpcore==1.0.9
+httpx==0.28.1
+huggingface-hub==0.33.4
+idna==3.10
+imageio==2.35.1
+importlib_metadata==8.5.0
+importlib_resources==6.4.5
+Jinja2==3.1.6
+kiwisolver==1.4.7
+lazy_loader==0.4
+loguru==0.7.3
+Markdown==3.7
+markdown-it-py==3.0.0
+MarkupSafe==2.1.5
+matplotlib==3.7.5
+mdurl==0.1.2
+networkx==3.1
+ninja==1.11.1.4
+numpy==1.24.4
+nvidia-pyindex==1.0.9
+oauthlib==3.3.1
+onnx==1.8.1
+onnx-simplifier==0.3.5
+onnxoptimizer==0.3.13
+onnxruntime==1.8.0
+opencv-python==4.5.5.64
+orjson==3.10.15
+packaging==25.0
+pandas==2.0.3
+pillow==10.4.0
+protobuf==5.29.5
+psutil==5.9.8
+pyasn1==0.6.1
+pyasn1_modules==0.4.2
+pycocotools==2.0.7
+pydantic==2.10.6
+pydantic_core==2.27.2
+pydub==0.25.1
+Pygments==2.19.2
+pyparsing==3.1.4
+python-dateutil==2.9.0.post0
+python-multipart==0.0.20
+pytz==2025.2
+PyWavelets==1.4.1
+PyYAML==6.0.2
+requests==2.32.4
+requests-oauthlib==2.0.0
+rich==14.0.0
+rsa==4.9.1
+ruff==0.12.3
+scikit-image==0.21.0
+scipy==1.10.1
+semantic-version==2.10.0
+shellingham==1.5.4
+shiboken2==5.15.2.1
+six==1.17.0
+sniffio==1.3.1
+spaces==0.37.1
+starlette==0.44.0
+tabulate==0.9.0
+tensorboard==2.14.0
+tensorboard-data-server==0.7.2
+thop==0.1.1.post2209072238
+tifffile==2023.7.10
+tomlkit==0.12.0
+torch==1.13.1+cu116
+torchaudio==0.13.1+cu116
+torchvision==0.14.1+cu116
+tqdm==4.67.1
+typer==0.16.0
+typing_extensions==4.13.2
+tzdata==2025.2
+urllib3==2.2.3
+uvicorn==0.33.0
+websockets==12.0
+Werkzeug==3.0.6
+yolox==0.3.0
+zipp==3.20.2