From beeb62025e9ab7aa5d513b5c86fa1fcbad9d9b15 Mon Sep 17 00:00:00 2001
From: psoubrie <Philippe.Soubrier@UGent.be>
Date: Mon, 8 Apr 2024 21:23:16 +0200
Subject: [PATCH] done

---
 .gitignore       |   8 ++++
 img2xlsx.py      | 121 +++++++++++++++++++++++++++++++++++++++++++++++
 readme.md        |  52 ++++++++++++++++++++
 requirements.txt |   2 +
 4 files changed, 183 insertions(+)
 create mode 100644 .gitignore
 create mode 100644 img2xlsx.py
 create mode 100644 readme.md
 create mode 100644 requirements.txt

diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..6322ac8
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,8 @@
+*.jpg
+*.xlsx
+cropped/
+output/
+*.pyd
+*.pyc
+*.pyo
+env/
diff --git a/img2xlsx.py b/img2xlsx.py
new file mode 100644
index 0000000..a134645
--- /dev/null
+++ b/img2xlsx.py
@@ -0,0 +1,121 @@
+from paddleocr import PaddleOCR
+import os
+import cv2
+import xlsxwriter
+
+
+def ocr_image(image_path, ocr):
+    result = ocr.ocr(image_path, cls=True)
+    return result[0][0][1][0]
+
+
+def cropUI(image_path):
+    # small opencv window to crop the image
+    image = cv2.imread(image_path)
+    r = cv2.selectROI(image)
+    cv2.destroyAllWindows()
+
+    return r
+
+
+def levelUI(image_path, region, initial_threshold=127):
+    # Callback function for the trackbar
+    def on_trackbar(val):
+        _, binary_image = cv2.threshold(gray_image, val, 255, cv2.THRESH_BINARY)
+        cv2.imshow("Binary Image", binary_image)
+
+    # Load and convert the image to grayscale
+    image = cv2.imread(image_path)
+    gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
+    gray_image = gray_image[
+        int(region[1]) : int(region[1] + region[3]),
+        int(region[0]) : int(region[0] + region[2]),
+    ]
+
+    # Create a window and a trackbar
+    cv2.namedWindow("Binary Image")
+    cv2.createTrackbar("Threshold", "Binary Image", initial_threshold, 255, on_trackbar)
+
+    # Initialize display
+    on_trackbar(initial_threshold)
+    cv2.waitKey(0)
+    cv2.destroyAllWindows()
+
+
+def cropImage(image_path, r, threshold, flip, cropped_folder):
+    # crop the image and save it
+    image = cv2.imread(image_path)
+    cropped = image[int(r[1]) : int(r[1] + r[3]), int(r[0]) : int(r[0] + r[2])]
+    gray = cv2.cvtColor(cropped, cv2.COLOR_BGR2GRAY)
+    gray = cv2.threshold(gray, threshold, 255, cv2.THRESH_BINARY)[1]
+
+    # save with new name
+    if flip == "y":
+        gray = cv2.flip(gray, 1)
+
+    # save in subfolder cropped
+    new_name = os.path.join(cropped_folder, image_path.split("/")[-1])
+    cv2.imwrite(new_name, gray)
+
+
+def data2excel(data):
+    # save the data in an excel file
+    fileName = "data.xlsx"
+    workbook = xlsxwriter.Workbook(fileName)
+    worksheet = workbook.add_worksheet()
+
+    # write the data
+    row = 0
+    for key, value in data.items():
+        worksheet.write(row, 0, key)
+        worksheet.write(row, 1, value)
+        try:
+            worksheet.write(row, 2, float(value[:6]))
+        except Exception as _:
+            pass
+        row += 1
+
+    workbook.close()
+
+
+def main():
+    # ocr settings
+    ocr = PaddleOCR(use_angle_cls=True, lang="en")
+
+    # *** start GUIs ***
+    images = [f for f in os.listdir(FOLDER) if f.endswith(".jpg")]
+    region = cropUI(os.path.join(FOLDER, images[0]))
+    threshold = levelUI(os.path.join(FOLDER, images[0]), region)
+
+    flip = input("Do you want to flip the images horizontaly? (y/n): ")
+
+    cropped_folder = os.path.join(FOLDER, "cropped")
+    if not os.path.exists(cropped_folder):
+        os.makedirs(cropped_folder)
+
+    # *** start cropping ***
+    for image in images:
+        cropImage(os.path.join(FOLDER, image), region, threshold, flip, cropped_folder)
+
+    # *** start OCR ***
+    cropped_images = [f for f in os.listdir(cropped_folder) if f.endswith(".jpg")]
+    data = {}
+    for image in cropped_images:
+        try:
+            path = os.path.join(cropped_folder, image)
+            text = ocr_image(path, ocr)
+        except Exception as _:
+            print("Error in cropped image")
+            continue
+
+        data[image] = text
+
+    # *** save data in excel ***
+    data2excel(data)
+
+    print("All images cropped successfully")
+
+
+if __name__ == "__main__":
+    FOLDER = "."
+    main()
diff --git a/readme.md b/readme.md
new file mode 100644
index 0000000..cb00f4e
--- /dev/null
+++ b/readme.md
@@ -0,0 +1,52 @@
+## Astral-sh UV
+
+### Getting started
+
+```bash
+# On macOS and Linux.
+curl -LsSf https://astral.sh/uv/install.sh | sh
+
+# On Windows.
+powershell -c "irm https://astral.sh/uv/install.ps1 | iex"
+
+# With pip.
+pip install uv
+```
+
+### Create virtual environment
+
+```bash
+uv venv  # Create a virtual environment at .venv.
+```
+
+### Activate virtual environment
+
+```bash
+# On macOS and Linux.
+source .venv/bin/activate
+
+# On Windows.
+.venv\Scripts\activate
+```
+
+# Install packages
+
+```bash
+uv pip install flask                # Install Flask.
+uv pip install -r requirements.txt  # Install from a requirements.txt file.
+uv pip install -e .                 # Install the current project in editable mode.
+uv pip install "package @ ."        # Install the current project from disk.
+uv pip install "flask[dotenv]"      # Install Flask with "dotenv" extra.
+```
+
+### Generate requirements.txt
+
+```bash
+uv pip freeze | uv pip compile - -o requirements.txt  # Lock the current environment.
+```
+
+### Update to requirements.txt
+
+```bash
+uv pip sync requirements.txt
+```
\ No newline at end of file
diff --git a/requirements.txt b/requirements.txt
new file mode 100644
index 0000000..f6cbdea
--- /dev/null
+++ b/requirements.txt
@@ -0,0 +1,2 @@
+paddlepaddle==2.6.0
+paddleocr>=2.0.1
\ No newline at end of file