ai文字加密图片生成器

# -*- coding: utf-8-*- # Author: Jack Cui import sys from turtle import width import cv2 from PIL import Image import numpy as np def parse_from_img(img_path):   img = cv2.imread(img_path)   img_h, img_w, _ = img.shape   stepx = 10   stepy = 10   sml_w = img_w // stepx   sml_h = img_h // stepy   res_img = np.zeros((sml_h, sml_w, 3), np.uint8)   for m in range(0, sml_w):     for n in range(0, sml_h):       map_col = int(m * stepx + stepx * 0.5)       map_row = int(n * stepy + stepy * 0.5)       res_img[n, m] = img[map_row, map_col]   return res_img def generate_img(big_img_path, small_img_path):   big_img = cv2.imread(big_img_path)   sml_img = cv2.imread(small_img_path)   dst_img = big_img.copy()   big_h, big_w, _ = big_img.shape   sml_h, sml_w, _ = sml_img.shape   stepx = big_w / sml_w   stepy = big_h / sml_h   for m in range(0, sml_w):     for n in range(0, sml_h):       map_col = int(m * stepx + stepx * 0.5)       map_row = int(n * stepy + stepy * 0.5)       if map_col < big_w and map_row < big_h:         dst_img[map_row, map_col] = sml_img[n, m]   return dst_img def Img2Text(img_fname):   img = cv2.imread(img_fname)   height, width, _ = img.shape   text_list = []   for h in range(height):     for w in range(width):       R, G, B = img[h, w]       if R | G | B == 0:         break       idx = (G << 8) + B       text_list.append(chr(idx))   text = "".join(text_list)   with open("斗破苍穹_dec.txt", "a", encoding="utf-8") as f:     f.write(text) def Text2Img(txt_fname, save_fname):   with open(txt_fname, "r", encoding="utf-8") as f:     text = f.read()   text_len = len(text)   img_w = 1000   img_h = 1680   img = np.zeros((img_h, img_w, 3))   x = 0   y = 0   for each_text in text:     idx = ord(each_text)     rgb = (0, (idx & 0xFF00) >> 8, idx & 0xFF)     img[y, x] = rgb     if x == img_w - 1:       x = 0       y += 1     else:       x += 1   cv2.imwrite(save_fname, img) def big_with_small(big_img_path, small_img_path, res_img_path):   """   大图里藏小图   """   dst_img = generate_img(big_img_path, small_img_path)   cv2.imwrite(res_img_path, dst_img) if __name__ == "__main__":   # 将小图藏于大图中，并保存结果   big_img_path = "1.png"   small_img_path = "2.png"   res_img_path = "res.png"   big_with_small(big_img_path, small_img_path, res_img_path)   # 从大图中，解析出小图   parsed_img_fname = "parsed_img.png"   parsed_img = parse_from_img(res_img_path)   cv2.imwrite(parsed_img_fname, parsed_img)   # 文本生成图片   txt_fname = "斗破苍穹.txt"   txt_img_fname = "dpcq.png"   txt_res_img_path = "text_res.png"   Text2Img(txt_fname, txt_img_fname)   # 将生成的文本图片，藏于大图中，并保存结果   big_with_small(big_img_path, txt_img_fname, txt_res_img_path)   # 从藏有文本的大图中，解析出文本小图   parsed_img_text_fname = "pares_text_img.png"   parsed_img_text = parse_from_img(txt_res_img_path)   cv2.imwrite(parsed_img_text_fname, parsed_img_text)   # 从文本图片中，解析出文字   Img2Text(parsed_img_text_fname)