Upload 11 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+weights/* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,10 +1,99 @@
 ---
-title: Blood Cell Detector
-emoji: 🏢
-colorFrom: gray
-colorTo: yellow
+title: Demo Cell Detection Gradio
+emoji: 📈
+colorFrom: indigo
+colorTo: indigo
 sdk: docker
 pinned: false
+license: gpl-3.0
+metrics:
+- accuracy
+- map
+tags:
+- yolo
+- cell detection
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Model Card for Automatic Identification and Counting of Blood Cells
+
+This model is designed for the automatic identification and counting of blood cells from smear images, using a machine learning approach.
+
+## Model Details
+
+### Model Description
+
+- **Developed by:** [Mahmudul Alam](https://github.com/MahmudulAlam)
+- **Shared by:** [Pablo Rocamora-Garcia](https://github.com/pablorocg)
+- **Model type:** Object Detection and Classification (YOLO)
+- **License:** GNU General Public License v3.0
+
+### Model Sources
+
+- **Repository:** https://github.com/MahmudulAlam/Automatic-Identification-and-Counting-of-Blood-Cells
+- **Paper:** [Machine learning approach of automatic identification and counting of blood cells](https://doi.org/10.1049/htl.2018.5098)
+- **Demo:** (Demo link or more information needed)
+
+## Uses
+
+### Direct Use
+
+This model is intended for use in medical diagnosis to evaluate overall health condition through the automatic identification and counting of red blood cells, white blood cells, and platelets from blood smear images.
+
+### Out-of-Scope Use
+
+The model is not intended for use outside of the scope of medical imaging and blood cell analysis. Misuse or application in other domains may result in inaccurate or irrelevant results.
+
+## How to Get Started with the Model
+
+To get started with the model, download the trained weights, set up the environment with TensorFlow and TF-slim, and run `detect.py`. Detailed instructions can be found in the [repository](https://github.com/MahmudulAlam/Automatic-Identification-and-Counting-of-Blood-Cells).
+
+## Training Details
+
+### Training Data
+
+The model was trained using the [Complete Blood Count (CBC) Dataset](https://github.com/MahmudulAlam/Complete-Blood-Cell-Count-Dataset), which contains images of blood smears for red blood cells, white blood cells, and platelets.
+
+### Results
+
+### Performance on Blood Cell Detection
+
+The performance of different CNN architectures with the YOLO algorithm for detecting different types of blood cells is summarized in the following table:
+
+| Model       | RBC Accuracy (%) | WBC Accuracy (%) | Platelet Accuracy (%) | mAP   | Execution Time (ms) |
+|-------------|------------------|------------------|-----------------------|-------|---------------------|
+| Tiny YOLO   | 96.09            | 86.89            | 96.36                 | 0.623 | 660                 |
+| VGG-16      | 72.98            | 100.00           | 90.91                 | 0.713 | 3106                |
+| ResNet50    | 79.80            | 95.08            | 87.27                 | 0.743 | 3711                |
+| InceptionV3 | 87.75            | 100.00           | 96.36                 | 0.682 | 2630                |
+| MobileNet   | 74.24            | 93.44            | 83.64                 | 0.520 | 784                 |
+
+### Accuracy of Counting Blood Cells
+
+The accuracy of counting Red Blood Cells (RBCs), White Blood Cells (WBCs), and platelets using the proposed method is detailed in the table below:
+
+| Cell Type   | Ground Truths | Estimated Count | Accuracy (%) |
+|-------------|---------------|-----------------|--------------|
+| RBCs        | 792           | 823             | 96.09        |
+| WBCs        | 65            | 53              | 86.89        |
+| Platelets   | 155           | 353             | 96.36        |
+
+## Citation
+
+**BibTeX:**
+
+```bibtex
+@article{https://doi.org/10.1049/htl.2018.5098,
+  title={Machine learning approach of automatic identification and counting of blood cells},
+  author={Alam, Mohammad Mahmudul and Islam, Mohammad Tariqul},
+  journal={Healthcare Technology Letters},
+  volume={6},
+  number={4},
+  pages={103-108},
+  year={2019},
+  doi={https://doi.org/10.1049/htl.2018.5098},
+  url={https://ietresearch.onlinelibrary.wiley.com/doi/abs/10.1049/htl.2018.5098}
+}
+```
+
+## Model Card Authors
+- [Pablo Rocamora-Garcia](https://github.com/pablorocg)

app.py

@@ -0,0 +1,48 @@
+import gradio as gr
+import pandas as pd
+import os
+import cv2
+# from inference import inference
+from PIL import Image, ImageFilter
+
+def prueba_inferencia(img):
+    
+    df = pd.DataFrame({"Cell type": ["rbc", "wbc", "platelets"], "Count": [28, 55, 43]})
+
+    # Aplicar filtro gaussiano
+    img = cv2.imread(img, cv2.IMREAD_COLOR)
+    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+
+
+    return [img, df]
+
+
+def run():
+    # Crear la interfaz de Gradio
+    interface = gr.Interface(
+        fn=prueba_inferencia,#inference
+        inputs=gr.Image(type="filepath"),
+        outputs=[gr.Image(type="pil"), 
+                gr.BarPlot(
+                    x="Cell type",
+                    y="Count",
+                    x_title="Cell type",
+                    y_title="Count",
+                    title="Cell Count",
+                    tooltip=["Cell type", "Count"],
+                    vertical=False,
+                    y_lim=[0, 100],
+                )],
+
+        examples=[os.path.join("./data", img) for img in os.listdir("./data")],
+        title="Detección de Células Sanguíneas",
+        description="Sube una imagen de microscopio de células sanguíneas para detectar RBC, WBC y platelets."
+    )
+
+    # Ejecutar la aplicación
+    interface.launch(server_name="0.0.0.0", server_port=7860) # Se encuentra en http://127.0.0.1:7860/
+    
+
+
+if __name__ == "__main__":
+    run()

flow

@@ -0,0 +1,6 @@
+#! /usr/bin/env python
+
+import sys
+from darkflow.cli import cliHandler
+
+cliHandler(sys.argv)

inference.py

@@ -0,0 +1,105 @@
+import cv2
+from utils import iou
+from scipy import spatial
+from darkflow.net.build import TFNet
+import pandas as pd
+
+def inference(file_name:str):
+    options = {'model': 'cfg/tiny-yolo-voc-3c.cfg',
+            'load': 3750,
+            'threshold': 0.1,
+            'gpu': 0.7}
+
+    tfnet = TFNet(options)
+
+    pred_bb = []  # predicted bounding box
+    pred_cls = []  # predicted class
+    pred_conf = []  # predicted class confidence
+
+    rbc = 0
+    wbc = 0
+    platelets = 0
+
+    cell = []
+    cls = []
+    conf = []
+
+    record = []
+    tl_ = []
+    br_ = []
+    iou_ = []
+    iou_value = 0
+
+    # tic = time.time()
+    image = cv2.imread(file_name)
+    output = tfnet.return_predict(image)
+
+    for prediction in output:
+        label = prediction['label']
+        confidence = prediction['confidence']
+        tl = (prediction['topleft']['x'], prediction['topleft']['y'])
+        br = (prediction['bottomright']['x'], prediction['bottomright']['y'])
+
+        if label == 'RBC' and confidence < .5:
+            continue
+        if label == 'WBC' and confidence < .25:
+            continue
+        if label == 'Platelets' and confidence < .25:
+            continue
+
+        # clearing up overlapped same platelets
+        if label == 'Platelets':
+            if record:
+                tree = spatial.cKDTree(record)
+                index = tree.query(tl)[1]
+                iou_value = iou(tl + br, tl_[index] + br_[index])
+                iou_.append(iou_value)
+
+            if iou_value > 0.1:
+                continue
+
+            record.append(tl)
+            tl_.append(tl)
+            br_.append(br)
+
+        center_x = int((tl[0] + br[0]) / 2)
+        center_y = int((tl[1] + br[1]) / 2)
+        center = (center_x, center_y)
+
+        if label == 'RBC':
+            color = (255, 0, 0)
+            rbc = rbc + 1
+        if label == 'WBC':
+            color = (0, 255, 0)
+            wbc = wbc + 1
+        if label == 'Platelets':
+            color = (0, 0, 255)
+            platelets = platelets + 1
+
+        radius = int((br[0] - tl[0]) / 2)
+        image = cv2.circle(image, center, radius, color, 2)
+        font = cv2.FONT_HERSHEY_COMPLEX
+        image = cv2.putText(image, label, (center_x - 15, center_y + 5), font, .5, color, 1)
+        cell.append([tl[0], tl[1], br[0], br[1]])
+
+        if label == 'RBC':
+            cls.append(0)
+        if label == 'WBC':
+            cls.append(1)
+        if label == 'Platelets':
+            cls.append(2)
+
+        conf.append(confidence)
+
+    # toc = time.time()
+    pred_bb.append(cell)
+    pred_cls.append(cls)
+    pred_conf.append(conf)
+    
+    df = pd.DataFrame({"Cell type": ["rbc", "wbc", "platelets"], "Count": [rbc, wbc, platelets]})
+
+    return [image, df]
+    
+
+
+

labels.txt

@@ -0,0 +1,3 @@
+WBC
+RBC
+Platelets

model_card.md

@@ -0,0 +1,83 @@
+# Model Card for Automatic Identification and Counting of Blood Cells
+
+This model is designed for the automatic identification and counting of blood cells from smear images, using a machine learning approach.
+
+## Model Details
+
+### Model Description
+
+- **Developed by:** Mahmudul Alam (https://github.com/MahmudulAlam)
+- **Shared by:** Pablo Rocamora-Garcia (https://github.com/pablorocg)
+- **Model type:** Object Detection and Classification (YOLO)
+- **License:** GNU General Public License v3.0
+
+### Model Sources
+
+- **Repository:** https://github.com/MahmudulAlam/Automatic-Identification-and-Counting-of-Blood-Cells
+- **Paper:** [Machine learning approach of automatic identification and counting of blood cells](https://doi.org/10.1049/htl.2018.5098)
+- **Demo:** (Demo link or more information needed)
+
+## Uses
+
+### Direct Use
+
+This model is intended for use in medical diagnosis to evaluate overall health condition through the automatic identification and counting of red blood cells, white blood cells, and platelets from blood smear images.
+
+### Out-of-Scope Use
+
+The model is not intended for use outside of the scope of medical imaging and blood cell analysis. Misuse or application in other domains may result in inaccurate or irrelevant results.
+
+## How to Get Started with the Model
+
+To get started with the model, download the trained weights, set up the environment with TensorFlow and TF-slim, and run `detect.py`. Detailed instructions can be found in the [repository](https://github.com/MahmudulAlam/Automatic-Identification-and-Counting-of-Blood-Cells).
+
+## Training Details
+
+### Training Data
+
+The model was trained using the [Complete Blood Count (CBC) Dataset](https://github.com/MahmudulAlam/Complete-Blood-Cell-Count-Dataset), which contains images of blood smears for red blood cells, white blood cells, and platelets.
+
+### Results
+
+### Performance on Blood Cell Detection
+
+The performance of different CNN architectures with the YOLO algorithm for detecting different types of blood cells is summarized in the following table:
+
+| Model       | RBC Accuracy (%) | WBC Accuracy (%) | Platelet Accuracy (%) | mAP   | Execution Time (ms) |
+|-------------|------------------|------------------|-----------------------|-------|---------------------|
+| Tiny YOLO   | 96.09            | 86.89            | 96.36                 | 0.623 | 660                 |
+| VGG-16      | 72.98            | 100.00           | 90.91                 | 0.713 | 3106                |
+| ResNet50    | 79.80            | 95.08            | 87.27                 | 0.743 | 3711                |
+| InceptionV3 | 87.75            | 100.00           | 96.36                 | 0.682 | 2630                |
+| MobileNet   | 74.24            | 93.44            | 83.64                 | 0.520 | 784                 |
+
+### Accuracy of Counting Blood Cells
+
+The accuracy of counting Red Blood Cells (RBCs), White Blood Cells (WBCs), and platelets using the proposed method is detailed in the table below:
+
+| Cell Type   | Ground Truths | Estimated Count | Accuracy (%) |
+|-------------|---------------|-----------------|--------------|
+| RBCs        | 792           | 823             | 96.09        |
+| WBCs        | 65            | 53              | 86.89        |
+| Platelets   | 155           | 353             | 96.36        |
+
+## Citation
+
+**BibTeX:**
+
+```bibtex
+@article{https://doi.org/10.1049/htl.2018.5098,
+  title={Machine learning approach of automatic identification and counting of blood cells},
+  author={Alam, Mohammad Mahmudul and Islam, Mohammad Tariqul},
+  journal={Healthcare Technology Letters},
+  volume={6},
+  number={4},
+  pages={103-108},
+  year={2019},
+  doi={https://doi.org/10.1049/htl.2018.5098},
+  url={https://ietresearch.onlinelibrary.wiley.com/doi/abs/10.1049/htl.2018.5098}
+}
+```
+
+## Model Card Authors
+- [Pablo Rocamora-Garcia](https://github.com/pablorocg)

requirements.txt

@@ -0,0 +1,5 @@
+opencv-python-headless==4.5.5.64
+tensorflow-gpu==2.2.0 
+tf-slim==1.1.0 
+Cython==0.29.2
+gradio==2.9.4

setup.py

@@ -0,0 +1,74 @@
+from setuptools import setup, find_packages
+from setuptools.extension import Extension
+from Cython.Build import cythonize
+import numpy
+import os
+import imp
+
+VERSION = imp.load_source('version', os.path.join('.', 'darkflow', 'version.py'))
+VERSION = VERSION.__version__
+
+if os.name == 'nt':
+    ext_modules = [
+        Extension("darkflow.cython_utils.nms",
+                  sources=["darkflow/cython_utils/nms.pyx"],
+                  # libraries=["m"] # Unix-like specific
+                  include_dirs=[numpy.get_include()]
+                  ),
+        Extension("darkflow.cython_utils.cy_yolo2_findboxes",
+                  sources=["darkflow/cython_utils/cy_yolo2_findboxes.pyx"],
+                  # libraries=["m"] # Unix-like specific
+                  include_dirs=[numpy.get_include()]
+                  ),
+        Extension("darkflow.cython_utils.cy_yolo_findboxes",
+                  sources=["darkflow/cython_utils/cy_yolo_findboxes.pyx"],
+                  # libraries=["m"] # Unix-like specific
+                  include_dirs=[numpy.get_include()]
+                  )
+    ]
+
+elif os.name == 'posix':
+    ext_modules = [
+        Extension("darkflow.cython_utils.nms",
+                  sources=["darkflow/cython_utils/nms.pyx"],
+                  libraries=["m"],  # Unix-like specific
+                  include_dirs=[numpy.get_include()]
+                  ),
+        Extension("darkflow.cython_utils.cy_yolo2_findboxes",
+                  sources=["darkflow/cython_utils/cy_yolo2_findboxes.pyx"],
+                  libraries=["m"],  # Unix-like specific
+                  include_dirs=[numpy.get_include()]
+                  ),
+        Extension("darkflow.cython_utils.cy_yolo_findboxes",
+                  sources=["darkflow/cython_utils/cy_yolo_findboxes.pyx"],
+                  libraries=["m"],  # Unix-like specific
+                  include_dirs=[numpy.get_include()]
+                  )
+    ]
+
+else:
+    ext_modules = [
+        Extension("darkflow.cython_utils.nms",
+                  sources=["darkflow/cython_utils/nms.pyx"],
+                  libraries=["m"]  # Unix-like specific
+                  ),
+        Extension("darkflow.cython_utils.cy_yolo2_findboxes",
+                  sources=["darkflow/cython_utils/cy_yolo2_findboxes.pyx"],
+                  libraries=["m"]  # Unix-like specific
+                  ),
+        Extension("darkflow.cython_utils.cy_yolo_findboxes",
+                  sources=["darkflow/cython_utils/cy_yolo_findboxes.pyx"],
+                  libraries=["m"]  # Unix-like specific
+                  )
+    ]
+
+setup(
+    version=VERSION,
+    name='darkflow',
+    description='Darkflow',
+    license='GPLv3',
+    url='https://github.com/thtrieu/darkflow',
+    packages=find_packages(),
+    scripts=['flow'],
+    ext_modules=cythonize(ext_modules)
+)

test.Dockerfile

@@ -0,0 +1,20 @@
+FROM python:3.6
+
+WORKDIR /code
+
+COPY ./requirements.txt /code/requirements.txt
+
+# RUN pip install --no-cache-dir -r /code/requirements.txt
+RUN pip install gradio==2.9.4
+RUN pip install pandas
+RUN pip install Cython==0.29.2
+RUN pip install tf-slim==1.1.0
+RUN pip install opencv-python-headless==4.5.5.64
+RUN pip install tensorflow-gpu==2.2.0 
+
+
+COPY . .
+
+EXPOSE 7860
+
+CMD ["python", "app.py"]

utils.py

@@ -0,0 +1,15 @@
+def iou(boxA, boxB):
+    xA = max(boxA[0], boxB[0])
+    yA = max(boxA[1], boxB[1])
+    xB = min(boxA[2], boxB[2])
+    yB = min(boxA[3], boxB[3])
+
+    interArea = (xB - xA + 1) * (yB - yA + 1)
+
+    if interArea < 0:
+        interArea = 0
+
+    boxAArea = (boxA[2] - boxA[0] + 1) * (boxA[3] - boxA[1] + 1)
+    boxBArea = (boxB[2] - boxB[0] + 1) * (boxB[3] - boxB[1] + 1)
+
+    return interArea / float(boxAArea + boxBArea - interArea)