image.py 13 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361
  1. # This Python file uses the following encoding: utf-8
  2. # @author runhey
  3. # github https://github.com/runhey
  4. import cv2
  5. import numpy as np
  6. from numpy import float32, int32, uint8, fromfile
  7. from pathlib import Path
  8. from atom.RuleImageMallResourceMixin import RuleImageMallResourceMixin
  9. from module.base.decorator import cached_property
  10. from module.logger import logger
  11. from module.base.utils import is_approx_rectangle
  12. class RuleImage(RuleImageMallResourceMixin):
  13. debug_mode: bool = False
  14. def __init__(self, roi_front: tuple, roi_back: tuple, method: str, threshold: float, file: str) -> None:
  15. """
  16. 初始化
  17. :param roi_front: 前置roi
  18. :param roi_back: 后置roi 用于匹配的区域
  19. :param method: 匹配方法 "Template matching"
  20. :param threshold: 阈值 0.8
  21. :param file: 相对路径, 带后缀
  22. """
  23. self._match_init = False # 这个是给后面的 等待图片稳定
  24. self._image = None # 这个是匹配的目标
  25. self._kp = None #
  26. self._des = None
  27. self.method = method
  28. self.roi_front: list = list(roi_front)
  29. self.roi_back = roi_back
  30. self.threshold = threshold
  31. self.file = file
  32. @cached_property
  33. def name(self) -> str:
  34. """
  35. :return:
  36. """
  37. return Path(self.file).stem.upper()
  38. def __str__(self):
  39. return self.name
  40. __repr__ = __str__
  41. def __eq__(self, other):
  42. return str(self) == str(other)
  43. def __hash__(self):
  44. return hash(self.name)
  45. def __bool__(self):
  46. return True
  47. def load_image(self) -> None:
  48. """
  49. 加载图片
  50. :return:
  51. """
  52. if self._image is not None:
  53. return
  54. img = cv2.imdecode(fromfile(self.file, dtype=uint8), -1)
  55. img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
  56. self._image = img
  57. height, width, channels = self._image.shape
  58. if height != self.roi_front[3] or width != self.roi_front[2]:
  59. self.roi_front[2] = width
  60. self.roi_front[3] = height
  61. logger.debug(f"{self.name} roi_front size changed to {width}x{height}")
  62. def load_kp_des(self) -> None:
  63. if self._kp is not None and self._des is not None:
  64. return
  65. self._kp, self._des = self.sift.detectAndCompute(self.image, None)
  66. @property
  67. def image(self):
  68. """
  69. 获取图片
  70. :return:
  71. """
  72. if self._image is None:
  73. self.load_image()
  74. return self._image
  75. @cached_property
  76. def is_template_match(self) -> bool:
  77. """
  78. 是否是模板匹配
  79. :return:
  80. """
  81. return self.method == "Template matching"
  82. @cached_property
  83. def is_sift_flann(self) -> bool:
  84. return self.method == "Sift Flann"
  85. @cached_property
  86. def sift(self):
  87. return cv2.SIFT_create()
  88. @cached_property
  89. def kp(self):
  90. if self._kp is None:
  91. self.load_kp_des()
  92. return self._kp
  93. @cached_property
  94. def des(self):
  95. if self._des is None:
  96. self.load_kp_des()
  97. return self._des
  98. def corp(self, image: np.array, roi: list = None) -> np.array:
  99. """
  100. 截取图片
  101. :param image:
  102. :param roi
  103. :return:
  104. """
  105. if roi is None:
  106. x, y, w, h = self.roi_back
  107. else:
  108. x, y, w, h = roi
  109. x, y, w, h = int(x), int(y), int(w), int(h)
  110. return image[y:y + h, x:x + w]
  111. def match(self, image: np.array, threshold: float = None) -> bool:
  112. """
  113. :param threshold:
  114. :param image:
  115. :return:
  116. """
  117. if threshold is None:
  118. threshold = self.threshold
  119. if not self.is_template_match:
  120. return self.sift_match(image)
  121. # raise Exception(f"unknown method {self.method}")
  122. source = self.corp(image)
  123. mat = self.image
  124. if mat is None or mat.shape[0] == 0 or mat.shape[1] == 0:
  125. logger.error(f"Template image is invalid: {mat.shape}") #检测模板尺寸,不合法则不进行匹配,避免两次截图画面完全相同造成模板不合法
  126. return True # 如果模板图像无效,直接返回 True
  127. res = cv2.matchTemplate(source, mat, cv2.TM_CCOEFF_NORMED)
  128. min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res) # 最小匹配度,最大匹配度,最小匹配度的坐标,最大匹配度的坐标
  129. if self.debug_mode:
  130. logger.attr(self.name, f'matching score {max_val:.5f}')
  131. if max_val > threshold:
  132. self.roi_front[0] = max_loc[0] + self.roi_back[0]
  133. self.roi_front[1] = max_loc[1] + self.roi_back[1]
  134. return True
  135. else:
  136. return False
  137. def match_all(self, image: np.array, threshold: float = None, roi: list = None) -> list[tuple]:
  138. """
  139. 区别于match,这个是返回所有的匹配结果
  140. :param roi:
  141. :param image:
  142. :param threshold:
  143. :return:
  144. """
  145. if roi is not None:
  146. self.roi_back = roi
  147. if threshold is None:
  148. threshold = self.threshold
  149. if not self.is_template_match:
  150. raise Exception(f"unknown method {self.method}")
  151. source = self.corp(image)
  152. mat = self.image
  153. results = cv2.matchTemplate(source, mat, cv2.TM_CCOEFF_NORMED)
  154. locations = np.where(results >= threshold)
  155. matches = []
  156. for pt in zip(*locations[::-1]): # (x, y) coordinates
  157. score = results[pt[1], pt[0]]
  158. # 得分, x, y, w, h
  159. x = self.roi_back[0] + pt[0]
  160. y = self.roi_back[1] + pt[1]
  161. matches.append((score, x, y, mat.shape[1], mat.shape[0]))
  162. return matches
  163. def match_all_any(self, image: np.array, threshold: float = None, roi: list = None, nms_threshold: float = 0.3) -> list[tuple]:
  164. """
  165. 区别于match,这个是返回所有的匹配结果,去除冗余匹配项(例如:多个框选区域重叠的情况)时使用。
  166. :param roi:
  167. :param image:
  168. :param threshold:
  169. :return:
  170. """
  171. if roi is not None:
  172. self.roi_back = roi
  173. if threshold is None:
  174. threshold = self.threshold
  175. if not self.is_template_match:
  176. raise Exception(f"unknown method {self.method}")
  177. source = self.corp(image)
  178. mat = self.image
  179. results = cv2.matchTemplate(source, mat, cv2.TM_CCOEFF_NORMED)
  180. locations = np.where(results >= threshold)
  181. matches = []
  182. for pt in zip(*locations[::-1]): # (x, y) coordinates
  183. score = results[pt[1], pt[0]]
  184. # 得分, x, y, w, h
  185. x = self.roi_back[0] + pt[0]
  186. y = self.roi_back[1] + pt[1]
  187. matches.append((score, x, y, mat.shape[1], mat.shape[0]))
  188. if len(matches) > 0:
  189. scores = np.array([m[0] for m in matches])
  190. boxes = np.array([[m[1], m[2], m[3], m[4]] for m in matches])
  191. # 使用OpenCV的NMSBoxes
  192. indices = cv2.dnn.NMSBoxes(boxes.tolist(), scores.tolist(), score_threshold=threshold, nms_threshold=nms_threshold)
  193. filtered_matches = [matches[i] for i in indices]
  194. return filtered_matches
  195. return matches
  196. def coord(self) -> tuple:
  197. """
  198. 获取roi_front的随机的点击的坐标
  199. :return:
  200. """
  201. x, y, w, h = self.roi_front
  202. return x + np.random.randint(0, w), y + np.random.randint(0, h)
  203. def coord_more(self) -> tuple:
  204. """
  205. 获取roi_back的随机的点击的坐标
  206. :return:
  207. """
  208. x, y, w, h = self.roi_back
  209. return x + np.random.randint(0, w), y + np.random.randint(0, h)
  210. def front_center(self) -> tuple:
  211. """
  212. 获取roi_front的中心坐标
  213. :return:
  214. """
  215. x, y, w, h = self.roi_front
  216. return int(x + w//2), int(y + h//2)
  217. def test_match(self, image: np.array):
  218. self.debug_mode = True
  219. if self.is_template_match:
  220. return self.match(image)
  221. if self.is_sift_flann:
  222. return self.sift_match(image, show=True)
  223. def sift_match(self, image: np.array, show=False) -> bool:
  224. """
  225. 特征匹配,同样会修改 roi_front
  226. :param image: 是游戏的截图,就是转通道后的截图
  227. :param show: 测试用的
  228. :return:
  229. """
  230. source = self.corp(image)
  231. kp, des = self.sift.detectAndCompute(source, None)
  232. # 参数1:index_params
  233. # 对于SIFT和SURF,可以传入参数index_params=dict(algorithm=FLANN_INDEX_KDTREE, trees=5)。
  234. # 对于ORB,可以传入参数index_params=dict(algorithm=FLANN_INDEX_LSH, table_number=6, key_size=12)。
  235. index_params = dict(algorithm=1, trees=5)
  236. # 参数2:search_params 指定递归遍历的次数,值越高结果越准确,但是消耗的时间也越多。
  237. search_params = dict(checks=50)
  238. # 根据设置的参数创建特征匹配器 指定匹配的算法和kd树的层数,指定返回的个数
  239. flann = cv2.FlannBasedMatcher(index_params, search_params)
  240. # 利用创建好的特征匹配器利用k近邻算法来用模板的特征描述符去匹配图像的特征描述符,k指的是返回前k个最匹配的特征区域
  241. # 返回的是最匹配的两个特征点的信息,返回的类型是一个列表,列表元素的类型是Dmatch数据类型,具体是什么我也不知道
  242. # 第一个参数是小图的des, 第二个参数是大图的des
  243. matches = flann.knnMatch(self.des, des, k=2)
  244. good = []
  245. result = True
  246. for i, (m, n) in enumerate(matches):
  247. # 设定阈值, 距离小于对方的距离的0.7倍我们认为是好的匹配点.
  248. if m.distance < 0.6 * n.distance:
  249. good.append(m)
  250. if len(good) >= 10:
  251. src_pts = float32([self.kp[m.queryIdx].pt for m in good]).reshape(-1, 1, 2)
  252. dst_pts = float32([kp[m.trainIdx].pt for m in good]).reshape(-1, 1, 2)
  253. # 计算透视变换矩阵m, 要求点的数量>=4
  254. m, mask = cv2.findHomography(src_pts, dst_pts, cv2.RANSAC, 5.0)
  255. # 创建一个包含模板图像四个角坐标的数组
  256. w, h = self.roi_front[2], self.roi_front[3]
  257. pts = float32([[0, 0], [0, h - 1], [w - 1, h - 1], [w - 1, 0]]).reshape(-1, 1, 2)
  258. if m is None:
  259. result = False
  260. else:
  261. dst = int32(cv2.perspectiveTransform(pts, m))
  262. self.roi_front[0] = dst[0, 0, 0] + self.roi_back[0]
  263. self.roi_front[1] = dst[0, 0, 1] + self.roi_back[1]
  264. if show:
  265. cv2.polylines(source, [dst], isClosed=True, color=(0, 0, 255), thickness=2)
  266. if not is_approx_rectangle(np.array([pos[0] for pos in dst])):
  267. result = False
  268. else:
  269. result = False
  270. # https://blog.csdn.net/cungudafa/article/details/105399278
  271. # https://blog.csdn.net/qq_45832961/article/details/122776322
  272. if show:
  273. # 准备一个空的掩膜来绘制好的匹配
  274. mask_matches = [[0, 0] for i in range(len(matches))]
  275. # 向掩膜中添加数据
  276. for i, (m, n) in enumerate(matches):
  277. if m.distance < 0.6 * n.distance: # 理论上0.7最好
  278. mask_matches[i] = [1, 0]
  279. img_matches = cv2.drawMatchesKnn(self.image, self.kp, source, kp, matches, None,
  280. matchColor=(0, 255, 0), singlePointColor=(255, 0, 0),
  281. matchesMask=mask_matches, flags=0)
  282. cv2.imshow(f'Sift Flann: {self.name}', img_matches)
  283. cv2.waitKey(0)
  284. cv2.destroyAllWindows()
  285. return result
  286. def match_mean_color(self, image, color: tuple, bias=10) -> bool:
  287. """
  288. :param image:
  289. :param color: rgb
  290. :param bias:
  291. :return:
  292. """
  293. image = self.corp(image)
  294. average_color = cv2.mean(image)
  295. # logger.info(f'{self.name} average_color: {average_color}')
  296. for i in range(3):
  297. if abs(average_color[i] - color[i]) > bias:
  298. return False
  299. return True
  300. if __name__ == "__main__":
  301. from dev_tools.assets_test import detect_image
  302. IMAGE_FILE = './log/test/QQ截图20240223151924.png'
  303. from tasks.Restart.assets import RestartAssets
  304. jade = RestartAssets.I_HARVEST_JADE
  305. jade.method = 'Sift Flann'
  306. sign = RestartAssets.I_HARVEST_SIGN
  307. sign.method = 'Sift Flann'
  308. print(jade.roi_front)
  309. detect_image(IMAGE_FILE, jade)
  310. detect_image(IMAGE_FILE, sign)
  311. print(jade.roi_front)