In [ ]:
!pip install tensorflow opencv-python


from google.colab import files
uploaded = files.upload()
# Upload:
#  - bangla_ocr_model.h5
#  - test_char.png (any single handwritten Bangla char image)

Requirement already satisfied: tensorflow in /usr/local/lib/python3.11/dist-packages (2.18.0)
Requirement already satisfied: opencv-python in /usr/local/lib/python3.11/dist-packages (4.12.0.88)
Requirement already satisfied: absl-py>=1.0.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (1.4.0)
Requirement already satisfied: astunparse>=1.6.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (1.6.3)
Requirement already satisfied: flatbuffers>=24.3.25 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (25.2.10)
Requirement already satisfied: gast!=0.5.0,!=0.5.1,!=0.5.2,>=0.2.1 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (0.6.0)
Requirement already satisfied: google-pasta>=0.1.1 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (0.2.0)
Requirement already satisfied: libclang>=13.0.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (18.1.1)
Requirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (3.4.0)
Requirement already satisfied: packaging in /usr/local/lib/python3.11/dist-packages (from tensorflow) (25.0)
Requirement already satisfied: protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<6.0.0dev,>=3.20.3 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (5.29.5)
Requirement already satisfied: requests<3,>=2.21.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (2.32.3)
Requirement already satisfied: setuptools in /usr/local/lib/python3.11/dist-packages (from tensorflow) (75.2.0)
Requirement already satisfied: six>=1.12.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (1.17.0)
Requirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (3.1.0)
Requirement already satisfied: typing-extensions>=3.6.6 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (4.14.1)
Requirement already satisfied: wrapt>=1.11.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (1.17.2)
Requirement already satisfied: grpcio<2.0,>=1.24.3 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (1.73.1)
Requirement already satisfied: tensorboard<2.19,>=2.18 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (2.18.0)
Requirement already satisfied: keras>=3.5.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (3.8.0)
Requirement already satisfied: numpy<2.1.0,>=1.26.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (2.0.2)
Requirement already satisfied: h5py>=3.11.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (3.14.0)
Requirement already satisfied: ml-dtypes<0.5.0,>=0.4.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (0.4.1)
Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (0.37.1)
Requirement already satisfied: wheel<1.0,>=0.23.0 in /usr/local/lib/python3.11/dist-packages (from astunparse>=1.6.0->tensorflow) (0.45.1)
Requirement already satisfied: rich in /usr/local/lib/python3.11/dist-packages (from keras>=3.5.0->tensorflow) (13.9.4)
Requirement already satisfied: namex in /usr/local/lib/python3.11/dist-packages (from keras>=3.5.0->tensorflow) (0.1.0)
Requirement already satisfied: optree in /usr/local/lib/python3.11/dist-packages (from keras>=3.5.0->tensorflow) (0.16.0)
Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/dist-packages (from requests<3,>=2.21.0->tensorflow) (3.4.2)
Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/dist-packages (from requests<3,>=2.21.0->tensorflow) (3.10)
Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/dist-packages (from requests<3,>=2.21.0->tensorflow) (2.5.0)
Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/dist-packages (from requests<3,>=2.21.0->tensorflow) (2025.7.14)
Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.11/dist-packages (from tensorboard<2.19,>=2.18->tensorflow) (3.8.2)
Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in /usr/local/lib/python3.11/dist-packages (from tensorboard<2.19,>=2.18->tensorflow) (0.7.2)
Requirement already satisfied: werkzeug>=1.0.1 in /usr/local/lib/python3.11/dist-packages (from tensorboard<2.19,>=2.18->tensorflow) (3.1.3)
Requirement already satisfied: MarkupSafe>=2.1.1 in /usr/local/lib/python3.11/dist-packages (from werkzeug>=1.0.1->tensorboard<2.19,>=2.18->tensorflow) (3.0.2)
Requirement already satisfied: markdown-it-py>=2.2.0 in /usr/local/lib/python3.11/dist-packages (from rich->keras>=3.5.0->tensorflow) (3.0.0)
Requirement already satisfied: pygments<3.0.0,>=2.13.0 in /usr/local/lib/python3.11/dist-packages (from rich->keras>=3.5.0->tensorflow) (2.19.2)
Requirement already satisfied: mdurl~=0.1 in /usr/local/lib/python3.11/dist-packages (from markdown-it-py>=2.2.0->rich->keras>=3.5.0->tensorflow) (0.1.2)
Upload widget is only available when the cell has been executed in the current browser session. Please rerun this cell to enable. 
Saving ko.jpeg to ko.jpeg
Saving bangla_ocr_model.h5 to bangla_ocr_model.h5
In [ ]:
from keras.models import load_model
model = load_model("bangla_ocr_model.h5")
print("✅ Model loaded.")

WARNING:absl:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.

✅ Model loaded.
In [ ]:
# 🔍 Load and predict image
import cv2
import numpy as np
from keras.preprocessing.image import img_to_array
import matplotlib.pyplot as plt

# 🔡 Put your original training label map here
class_names = ['অ', 'আ', 'ই', 'ঈ', 'উ', 'ঊ', 'ঋ', 'এ', 'ঐ', 'ও', 'ঔ',
               'ক', 'খ', 'গ', 'ঘ', 'ঙ', 'চ', 'ছ', 'জ', 'ঝ', 'ঞ',
               'ট', 'ঠ', 'ড', 'ঢ', 'ণ', 'ত', 'থ', 'দ', 'ধ', 'ন',
               'প', 'ফ', 'ব', 'ভ', 'ম', 'য', 'র', 'ল', 'শ', 'ষ',
               'স', 'হ', 'ড়', 'ঢ়', 'য়', 'ৎ', '০', '১', '২', '৩',
               '৪', '৫', '৬', '৭', '৮', '৯']

# Change to your image filename
img_path = "ko.jpeg"

img = cv2.imread(img_path, cv2.IMREAD_GRAYSCALE)
img = cv2.resize(img, (64, 64))
img = img_to_array(img).reshape(1, 64, 64, 1) / 255.0
pred = model.predict(img)[0]
predicted_label = class_names[np.argmax(pred)]

print(" Predicted character:", predicted_label)

# 🖼 Show the image
plt.imshow(cv2.imread(img_path), cmap='gray')
plt.title(f"Prediction: {predicted_label}")
plt.axis('off')
plt.show()

1/1 ━━━━━━━━━━━━━━━━━━━━ 0s 228ms/step
 Predicted character: ঈ

/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Glyph 2440 (\N{BENGALI LETTER II}) missing from font(s) DejaVu Sans.
  fig.canvas.print_figure(bytes_io, **kw)
/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Matplotlib currently does not support Bengali natively.
  fig.canvas.print_figure(bytes_io, **kw)
No description has been provided for this image
In [ ]:
# Upload image from local machine
from google.colab import files
uploaded = files.upload()

# Get the uploaded filename
img_path = list(uploaded.keys())[0]

# Load and predict image
import cv2
import numpy as np
from keras.preprocessing.image import img_to_array
import matplotlib.pyplot as plt

# Class label map (update if yours is different)
class_names = ['অ', 'আ', 'ই', 'ঈ', 'উ', 'ঊ', 'ঋ', 'এ', 'ঐ', 'ও', 'ঔ',
               'ক', 'খ', 'গ', 'ঘ', 'ঙ', 'চ', 'ছ', 'জ', 'ঝ', 'ঞ',
               'ট', 'ঠ', 'ড', 'ঢ', 'ণ', 'ত', 'থ', 'দ', 'ধ', 'ন',
               'প', 'ফ', 'ব', 'ভ', 'ম', 'য', 'র', 'ল', 'শ', 'ষ',
               'স', 'হ', 'ড়', 'ঢ়', 'য়', 'ৎ', '০', '১', '২', '৩',
               '৪', '৫', '৬', '৭', '৮', '৯']

img = cv2.imread(img_path, cv2.IMREAD_GRAYSCALE)
img = cv2.resize(img, (64, 64))
img = img_to_array(img).reshape(1, 64, 64, 1) / 255.0

pred = model.predict(img)[0]
print(np.argmax(pred))
predicted_label = class_names[np.argmax(pred)]

print("Predicted character:", predicted_label)

plt.imshow(cv2.imread(img_path), cmap='gray')
plt.title(f"Prediction: {predicted_label}")
plt.axis('off')
plt.show()
Upload widget is only available when the cell has been executed in the current browser session. Please rerun this cell to enable. 
Saving 20250724_150736.jpg to 20250724_150736.jpg
1/1 ━━━━━━━━━━━━━━━━━━━━ 0s 50ms/step
29
Predicted character: ধ
No description has been provided for this image
In [ ]:
import cv2
import numpy as np
from keras.preprocessing.image import img_to_array

def preprocess_stylized_image(img_path):
    img = cv2.imread(img_path, cv2.IMREAD_GRAYSCALE)

    # Threshold + invert
    _, thresh = cv2.threshold(img, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)

    # Crop bounding box
    coords = cv2.findNonZero(thresh)
    x, y, w, h = cv2.boundingRect(coords)
    cropped = thresh[y:y+h, x:x+w]

    # Resize while keeping aspect ratio and padding
    size = 64
    h, w = cropped.shape
    scale = size / max(h, w)
    resized = cv2.resize(cropped, (int(w * scale), int(h * scale)))

    # Create square canvas and center
    square = np.ones((size, size), dtype=np.uint8) * 0  # black background
    x_offset = (size - resized.shape[1]) // 2
    y_offset = (size - resized.shape[0]) // 2
    square[y_offset:y_offset+resized.shape[0], x_offset:x_offset+resized.shape[1]] = resized

    # Normalize and reshape
    img_array = img_to_array(square).reshape(1, 64, 64, 1) / 255.0
    return img_array
In [ ]:
img = preprocess_stylized_image("ko.jpeg")
pred = model.predict(img)[0]
print("Predicted character:", class_names[np.argmax(pred)])

1/1 ━━━━━━━━━━━━━━━━━━━━ 0s 39ms/step
Predicted character: ধ
In [ ]:
import cv2
import numpy as np
from keras.preprocessing.image import img_to_array
from tensorflow.keras.models import load_model
import matplotlib.pyplot as plt
from google.colab import files  # remove this line if using Jupyter



# Class names
class_names = ['অ', 'আ', 'ই', 'ঈ', 'উ', 'ঊ', 'ঋ', 'এ', 'ঐ', 'ও', 'ঔ',
               'ক', 'খ', 'গ', 'ঘ', 'ঙ', 'চ', 'ছ', 'জ', 'ঝ', 'ঞ',
               'ট', 'ঠ', 'ড', 'ঢ', 'ণ', 'ত', 'থ', 'দ', 'ধ', 'ন',
               'প', 'ফ', 'ব', 'ভ', 'ম', 'য', 'র', 'ল', 'শ', 'ষ',
               'স', 'হ', 'ড়', 'ঢ়', 'য়', 'ৎ', '০', '১', '২', '৩',
               '৪', '৫', '৬', '৭', '৮', '৯']

# Preprocessing function
def preprocess_stylized_image(img_path):
    img = cv2.imread(img_path, cv2.IMREAD_GRAYSCALE)
    _, thresh = cv2.threshold(img, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)
    coords = cv2.findNonZero(thresh)
    x, y, w, h = cv2.boundingRect(coords)
    cropped = thresh[y:y+h, x:x+w]
    size = 64
    h, w = cropped.shape
    scale = size / max(h, w)
    resized = cv2.resize(cropped, (int(w * scale), int(h * scale)))
    square = np.ones((size, size), dtype=np.uint8) * 0
    x_offset = (size - resized.shape[1]) // 2
    y_offset = (size - resized.shape[0]) // 2
    square[y_offset:y_offset+resized.shape[0], x_offset:x_offset+resized.shape[1]] = resized
    img_array = img_to_array(square).reshape(1, 64, 64, 1) / 255.0
    return img_array, square

# Upload and predict
uploaded = files.upload()
img_path = next(iter(uploaded))

img_array, display_img = preprocess_stylized_image(img_path)
pred = model.predict(img_array)[0]
predicted_label = class_names[np.argmax(pred)]

print("Predicted character:", predicted_label)

plt.imshow(display_img, cmap='gray')
plt.title(f"Prediction: {predicted_label}")
plt.axis('off')
plt.show()
Upload widget is only available when the cell has been executed in the current browser session. Please rerun this cell to enable. 
Saving images.png to images (3).png
1/1 ━━━━━━━━━━━━━━━━━━━━ 0s 39ms/step
Predicted character: ঈ

/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Glyph 2440 (\N{BENGALI LETTER II}) missing from font(s) DejaVu Sans.
  fig.canvas.print_figure(bytes_io, **kw)
No description has been provided for this image
In [ ]:
# STEP 1: Install dependencies
!pip install tensorflow opencv-python-headless --quiet

# STEP 2: Upload model + line image
from google.colab import files
uploaded = files.upload()  # Upload both model.h5 and line image

# Identify uploaded model and image file

img_file = [f for f in uploaded if not f.endswith('.h5')][0]

# STEP 3: Load model
from tensorflow.keras.models import load_model


# STEP 4: Character label list (60 total)
class_names = ['অ', 'আ', 'ই', 'ঈ', 'উ', 'ঊ', 'ঋ', 'এ', 'ঐ', 'ও', 'ঔ',
               'ক', 'খ', 'গ', 'ঘ', 'ঙ', 'চ', 'ছ', 'জ', 'ঝ', 'ঞ',
               'ট', 'ঠ', 'ড', 'ঢ', 'ণ', 'ত', 'থ', 'দ', 'ধ', 'ন',
               'প', 'ফ', 'ব', 'ভ', 'ম', 'য', 'র', 'ল', 'শ', 'ষ',
               'স', 'হ', 'ড়', 'ঢ়', 'য়', 'ৎ', '০', '১', '২', '৩',
               '৪', '৫', '৬', '৭', '৮', '৯']

# Auto-adjust class_names to model output
import numpy as np
from keras.preprocessing.image import img_to_array
import cv2
import matplotlib.pyplot as plt

model_output_classes = model.output_shape[-1]
if len(class_names) != model_output_classes:
    class_names = class_names[:model_output_classes]

# STEP 5: Predict function
def predict_line(img_path):
    img = cv2.imread(img_path, cv2.IMREAD_GRAYSCALE)
    if img is None:
        print(f"Error: Could not load image from {img_path}")
        return

    _, thresh = cv2.threshold(img, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)
    contours, _ = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
    bounding_boxes = [cv2.boundingRect(c) for c in contours]
    if not bounding_boxes:
        print("No characters detected.")
        plt.imshow(cv2.imread(img_path), cmap='gray')
        plt.title("No characters detected")
        plt.axis('off')
        plt.show()
        return

    bounding_boxes = sorted(bounding_boxes, key=lambda x: x[0])  # Left to right

    predicted_text = ""
    for box in bounding_boxes:
        x, y, w, h = box
        roi = thresh[y:y+h, x:x+w]
        size = 64
        h, w = roi.shape
        scale = size / max(h, w)
        resized = cv2.resize(roi, (int(w * scale), int(h * scale)))
        square = np.zeros((size, size), dtype=np.uint8)
        x_offset = (size - resized.shape[1]) // 2
        y_offset = (size - resized.shape[0]) // 2
        square[y_offset:y_offset+resized.shape[0], x_offset:x_offset+resized.shape[1]] = resized
        img_array = img_to_array(square).reshape(1, 64, 64, 1) / 255.0

        pred = model.predict(img_array, verbose=0)[0]
        label = class_names[np.argmax(pred)]
        predicted_text += label

    print("Predicted Line:", predicted_text)
    plt.imshow(cv2.imread(img_path), cmap='gray')
    plt.axis('off')
    plt.title(f"Predicted: {predicted_text}")
    plt.show()

# Run it
predict_line(img_file)
Upload widget is only available when the cell has been executed in the current browser session. Please rerun this cell to enable. 
Saving maxresdefault.jpg to maxresdefault (4).jpg

---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
/tmp/ipython-input-21-2206923193.py in <cell line: 0>()
     77 
     78 # Run it
---> 79 predict_line(img_file)

/tmp/ipython-input-21-2206923193.py in predict_line(img_path)
     67 
     68         pred = model.predict(img_array, verbose=0)[0]
---> 69         label = class_names[np.argmax(pred)]
     70         predicted_text += label
     71 

IndexError: list index out of range
In [ ]:
# STEP 1: Install dependencies
!pip install tensorflow opencv-python-headless --quiet

# STEP 2: Upload only the line image
from google.colab import files
uploaded = files.upload()  # Just upload image (not model)

img_file = list(uploaded.keys())[0]  # Grab the image filename

# STEP 3: Load pre-existing model (already in Colab runtime)
from tensorflow.keras.models import load_model
model = load_model('bangla_ocr_model.h5')  # Use exact filename if different

# STEP 4: Character label list (adjust if your model has fewer outputs)
class_names = ['অ', 'আ', 'ই', 'ঈ', 'উ', 'ঊ', 'ঋ', 'এ', 'ঐ', 'ও', 'ঔ',
               'ক', 'খ', 'গ', 'ঘ', 'ঙ', 'চ', 'ছ', 'জ', 'ঝ', 'ঞ',
               'ট', 'ঠ', 'ড', 'ঢ', 'ণ', 'ত', 'থ', 'দ', 'ধ', 'ন',
               'প', 'ফ', 'ব', 'ভ', 'ম', 'য', 'র', 'ল', 'শ', 'ষ',
               'স', 'হ', 'ড়', 'ঢ়', 'য়', 'ৎ', '০', '১', '২', '৩',
               '৪', '৫', '৬', '৭', '৮', '৯']

# Auto-adjust to model output size
import numpy as np
from keras.preprocessing.image import img_to_array
import cv2
import matplotlib.pyplot as plt

output_dim = model.output_shape[-1]
if len(class_names) != output_dim:
    class_names = class_names[:output_dim]

# STEP 5: Prediction function
def predict_line(img_path):
    img = cv2.imread(img_path, cv2.IMREAD_GRAYSCALE)
    if img is None:
        print(f"Error loading image: {img_path}")
        return

    _, thresh = cv2.threshold(img, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)
    contours, _ = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
    boxes = sorted([cv2.boundingRect(c) for c in contours], key=lambda b: b[0])

    predicted_text = ""
    for x, y, w, h in boxes:
        roi = thresh[y:y+h, x:x+w]
        size = 64
        scale = size / max(w, h)
        resized = cv2.resize(roi, (int(w * scale), int(h * scale)))
        square = np.zeros((size, size), dtype=np.uint8)
        x_offset = (size - resized.shape[1]) // 2
        y_offset = (size - resized.shape[0]) // 2
        square[y_offset:y_offset+resized.shape[0], x_offset:x_offset+resized.shape[1]] = resized

        img_array = img_to_array(square).reshape(1, 64, 64, 1) / 255.0
        pred = model.predict(img_array, verbose=0)[0]
        label_idx = np.argmax(pred)

        if label_idx < len(class_names):
            predicted_text += class_names[label_idx]
        else:
            predicted_text += '?'

    print("Predicted Line:", predicted_text)
    plt.imshow(cv2.imread(img_path), cmap='gray')
    plt.title(f"Prediction: {predicted_text}")
    plt.axis('off')
    plt.show()

# STEP 6: Run
predict_line(img_file)
Upload widget is only available when the cell has been executed in the current browser session. Please rerun this cell to enable. 
WARNING:absl:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.

Saving maxresdefault.jpg to maxresdefault (5).jpg
Predicted Line: ?১ঙৎৎৎঐৎথঙঈটৎঐৎঙ৭ঙ৭থৎথজৎরঈটথথৎ৭টঈ

/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Glyph 2535 (\N{BENGALI DIGIT ONE}) missing from font(s) DejaVu Sans.
  fig.canvas.print_figure(bytes_io, **kw)
/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Matplotlib currently does not support Bengali natively.
  fig.canvas.print_figure(bytes_io, **kw)
/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Glyph 2457 (\N{BENGALI LETTER NGA}) missing from font(s) DejaVu Sans.
  fig.canvas.print_figure(bytes_io, **kw)
/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Glyph 2510 (\N{BENGALI LETTER KHANDA TA}) missing from font(s) DejaVu Sans.
  fig.canvas.print_figure(bytes_io, **kw)
/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Glyph 2448 (\N{BENGALI LETTER AI}) missing from font(s) DejaVu Sans.
  fig.canvas.print_figure(bytes_io, **kw)
/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Glyph 2469 (\N{BENGALI LETTER THA}) missing from font(s) DejaVu Sans.
  fig.canvas.print_figure(bytes_io, **kw)
/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Glyph 2440 (\N{BENGALI LETTER II}) missing from font(s) DejaVu Sans.
  fig.canvas.print_figure(bytes_io, **kw)
/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Glyph 2463 (\N{BENGALI LETTER TTA}) missing from font(s) DejaVu Sans.
  fig.canvas.print_figure(bytes_io, **kw)
/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Glyph 2541 (\N{BENGALI DIGIT SEVEN}) missing from font(s) DejaVu Sans.
  fig.canvas.print_figure(bytes_io, **kw)
/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Glyph 2460 (\N{BENGALI LETTER JA}) missing from font(s) DejaVu Sans.
  fig.canvas.print_figure(bytes_io, **kw)
/usr/local/lib/python3.11/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Glyph 2480 (\N{BENGALI LETTER RA}) missing from font(s) DejaVu Sans.
  fig.canvas.print_figure(bytes_io, **kw)
No description has been provided for this image
In [ ]:
# ✅ STEP 1: Install dependencies
!pip install easyocr pix2tex transformers matplotlib -q
!apt install tesseract-ocr -y > /dev/null

# ✅ STEP 2: Upload image
from google.colab import files
uploaded = files.upload()
img_path = list(uploaded.keys())[0]

# ✅ STEP 3: Bangla OCR using EasyOCR
import easyocr
reader = easyocr.Reader(['bn'], gpu=False)
bangla_results = reader.readtext(img_path, detail=0)
bangla_text = ' '.join(bangla_results)

# ✅ STEP 4: Equation OCR using pix2tex
import torch
from PIL import Image
from pix2tex.cli import LatexOCR

ocr = LatexOCR()
eq_img = Image.open(img_path)
latex_result = ocr(eq_img)

# ✅ STEP 5: Display outputs
from IPython.display import display, Markdown

print("🔡 Bangla Text:\n", bangla_text)
print("\n🧮 LaTeX Equation:\n", latex_result)

# Render LaTeX inside notebook
display(Markdown(f"**📖 Rendered Equation:**  \n\n`{latex_result}`"))

WARNING: apt does not have a stable CLI interface. Use with caution in scripts.
Upload widget is only available when the cell has been executed in the current browser session. Please rerun this cell to enable. 
Saving Screenshot 2025-07-24 160701.jpg to Screenshot 2025-07-24 160701.jpg
🔡 Bangla Text:
 সরলরেখাটি দ্বারা উভয় অক্ষের ছেদাংশ = 2 এবং ছেদবিন্দু (2, 0) ও (0, 2) তাহলে, (1) নং বৃত্ত ও (11) ন? সরলরেখা দ্বারা আবদধ ক্ষেত্রের শ এর সীমা হবে & = [ থেকে $ = 2 পর্যন্ত এবন, ৮2 +১2 = 4 বা, ঢ2 = 4 - ভ2 বা, ড =   4 - ষ- [যেহেতু শ অক্ষের উপর ধনাত্মক] এবং ঢ = 3 - & [(ু1) নং হতে] ধরি, ঢ1 = ' 4 - স2 এবং ঢ2 = 2 - & (1) নং বৃত্ত ও (11) নং রেখা দ্বারা আবদ্ধ ক্ষুদ্রতর অংশের ক্ষেত্রফল অর্থাৎ চিত্রের ছায়াঘেরা অংশের ক্ষেত্রফল, অর্থাৎ চিত্রের ছায়াঘেরা অংশের ক্ষেত্রফল  (1 -92) ধ&

🧮 LaTeX Equation:
 \scriptstyle{\frac{\mu(m)b^{0}b^{0}}{\hbar\hbar^{0}}}

📖 Rendered Equation:

\scriptstyle{\frac{\mu(m)b^{0}b^{0}}{\hbar\hbar^{0}}}

In [ ]:
!pip uninstall -y docx
!pip install python-docx

Found existing installation: docx 0.2.4
Uninstalling docx-0.2.4:
  Successfully uninstalled docx-0.2.4
Collecting python-docx
  Downloading python_docx-1.2.0-py3-none-any.whl.metadata (2.0 kB)
Requirement already satisfied: lxml>=3.1.0 in /usr/local/lib/python3.11/dist-packages (from python-docx) (5.4.0)
Requirement already satisfied: typing_extensions>=4.9.0 in /usr/local/lib/python3.11/dist-packages (from python-docx) (4.14.1)
Downloading python_docx-1.2.0-py3-none-any.whl (252 kB)
   ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 253.0/253.0 kB 4.3 MB/s eta 0:00:00
Installing collected packages: python-docx
Successfully installed python-docx-1.2.0
In [ ]:
from docx import Document

doc = Document()
doc.add_heading('Bangla OCR + Math Equation Result', level=1)

doc.add_heading('📌 Bangla Text:', level=2)
doc.add_paragraph(bangla_text)

doc.add_heading('📌 LaTeX Equation:', level=2)
doc.add_paragraph(latex_result)

doc.save("ocr_result.docx")

from google.colab import files
files.download("ocr_result.docx")
In [ ]:
# @title 🔄 LaTeX to MathJax Converter

def latex_to_mathjax_inline(latex_str):
    return f"\\({latex_str}\\)"  # Inline math

def latex_to_mathjax_block(latex_str):
    return f"$$ {latex_str} $$"  # Block-level math

def sanitize_latex(latex_str):
    import re
    latex_str = re.sub(r'\\displaystyle', '', latex_str)
    return latex_str.strip()
In [ ]:
latex = r"9x + 27 = 49 + x - 2 + 14\sqrt{x - 2}"

cleaned = sanitize_latex(latex)
inline_mj = latex_to_mathjax_inline(cleaned)
block_mj = latex_to_mathjax_block(cleaned)

print("Inline MathJax:\n", inline_mj)
print("Block MathJax:\n", block_mj)

Inline MathJax:
 \(9x + 27 = 49 + x - 2 + 14\sqrt{x - 2}\)
Block MathJax:
 $$ 9x + 27 = 49 + x - 2 + 14\sqrt{x - 2} $$
In [ ]:
html_template = f"""
<!DOCTYPE html>
<html>
<head>
  <script src="https://polyfill.io/v3/polyfill.min.js?features=es6"></script>
  <script id="MathJax-script" async
    src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-mml-chtml.js">
  </script>
</head>
<body>
  <h3>Rendered Math (Block):</h3>
  <p>{block_mj}</p>
  <hr>
  <h3>Rendered Math (Inline):</h3>
  <p>This is inline math: {inline_mj}</p>
</body>
</html>
"""

# Save to file
with open("mathjax_preview.html", "w") as f:
    f.write(html_template)

# Download link
from google.colab import files
files.download("mathjax_preview.html")