Preprocess/preprocessImg.py

import cv2
import numpy as np
from PIL import Image, ImageEnhance


def PreprocessImg(image):
    if image is None:
        raise ValueError("尚未上傳圖片！")

    # 方案一
    pre_img_001 = preprocess_image001(image)
    # 方案二
    pre_img_002 = preprocess_image002(image)
    # 方案三
    pre_img_003 = preprocess_image003(image)
    # 方案四
    pre_img_004 = preprocess_image004(image)
    # 方案五
    pre_img_005 = preprocess_image005(image)

    return pre_img_001, pre_img_002, pre_img_003, pre_img_004, pre_img_005


# 方案一
def preprocess_image001(image):
    # 將影像轉換為 NumPy 數組
    image = np.array(image)
    # 轉為灰階影像
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    # 調整對比度
    enhancer = ImageEnhance.Contrast(Image.fromarray(gray))
    enhanced_image = enhancer.enhance(2)
    # 二值化
    _, binary = cv2.threshold(np.array(enhanced_image), 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
    # 去雜訊
    denoised = cv2.fastNlMeansDenoising(binary, None, 30, 7, 21)
    return Image.fromarray(denoised)


# 方案二
def preprocess_image002(image):
    # 將 PIL Image 轉換為 numpy array
    image_np = np.array(image)
    # 使用 OpenCV 進行預處理
    gray = cv2.cvtColor(image_np, cv2.COLOR_BGR2GRAY)  # 灰階化
    gray = cv2.bilateralFilter(gray, 11, 17, 17)  # 雙邊濾波去噪
    edged = cv2.Canny(gray, 30, 200)  # 邊緣檢測
    return Image.fromarray(edged)


# 方案三：自適應門檻和形態學變換
def preprocess_image003(image):
    # 將 PIL Image 轉換為 numpy array
    image_np = np.array(image)
    # 轉為灰階影像
    gray = cv2.cvtColor(image_np, cv2.COLOR_BGR2GRAY)
    # 自適應門檻處理
    adaptive_thresh = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, 11, 2)
    # 形態學變換 (開運算) 去除小噪點
    kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (3, 3))
    morph = cv2.morphologyEx(adaptive_thresh, cv2.MORPH_OPEN, kernel)
    return Image.fromarray(morph)


# 方案四：CLAHE（限制對比度自適應直方圖均衡）
def preprocess_image004(image):
    # 將 PIL Image 轉換為 numpy array
    image_np = np.array(image)
    # 轉為灰階影像
    gray = cv2.cvtColor(image_np, cv2.COLOR_BGR2GRAY)
    # 應用 CLAHE
    clahe = cv2.createCLAHE(clipLimit=2.0, tileGridSize=(8, 8))
    clahe_image = clahe.apply(gray)
    # 二值化
    _, binary = cv2.threshold(clahe_image, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
    return Image.fromarray(binary)


# 方案五：直方圖均衡化和高斯模糊
def preprocess_image005(image):
    # 將 PIL Image 轉換為 numpy array
    image_np = np.array(image)
    # 轉為灰階影像
    gray = cv2.cvtColor(image_np, cv2.COLOR_BGR2GRAY)
    # 直方圖均衡化
    equalized = cv2.equalizeHist(gray)
    # 高斯模糊
    blurred = cv2.GaussianBlur(equalized, (5, 5), 0)
    # 二值化
    _, binary = cv2.threshold(blurred, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
    return Image.fromarray(binary)


# 方案六：自適應去噪與銳化
def preprocess_image006(image):
    # 將 PIL Image 轉換為 numpy array
    image_np = np.array(image)
    # 轉為灰階影像
    gray = cv2.cvtColor(image_np, cv2.COLOR_BGR2GRAY)
    # 自適應去噪
    denoised = cv2.fastNlMeansDenoising(gray, None, 30, 7, 21)
    # 銳化
    kernel = np.array([[0, -1, 0], [-1, 5, -1], [0, -1, 0]])
    sharpened = cv2.filter2D(denoised, -1, kernel)
    # 二值化
    _, binary = cv2.threshold(sharpened, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
    return Image.fromarray(binary)