forked from Mufanc/Genshin-SmartFishingRod
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
新增 | 全局配置文件 新增 | 支持自定义游戏窗口标题和类名 新增 | 支持窗口裁剪 新增 | 支持快速切换模板 新增 | 现在可以自定义进度条的阈值 优化 | 改善代码组织结构和变量命名 优化 | 将耗时操作移动到单独的进程，大幅提高运行效率 优化 | 改变显示和隐藏叠加层的实现方式
- Loading branch information
Showing
28 changed files
with
534 additions
and
471 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,4 +1,9 @@ | ||
| # 虚拟环境 | ||
| /venv/ | ||
| /detects/clips/* | ||
|
|
||
| # 打包产生的文件 | ||
| /build/ | ||
| /dist/ | ||
| /dist/ | ||
|
|
||
| # 临时截图 | ||
| /detects/*/clips/* |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,4 @@ | ||
| from .detector import Detector | ||
| from .genshin import Genshin | ||
| from .hotkey import Hotkey | ||
| from .overlay import Overlay |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,130 @@ | ||
| import cv2 | ||
| import numpy as np | ||
| import yaml | ||
| from loguru import logger | ||
|
|
||
| from configs import configs | ||
|
|
||
| COLOR_RED = (0, 0, 255) | ||
| COLOR_GREEN = (0, 255, 0) | ||
| COLOR_BLUE = (255, 0, 0) | ||
| COLOR_YELLOW = (0, 255, 255) | ||
| COLOR_MAGENTA = (255, 0, 255) | ||
|
|
||
|
|
||
| class Detector(object): | ||
| def __init__(self): | ||
| self.font = cv2.FONT_HERSHEY_COMPLEX | ||
| self.model_name = configs["use-model"] | ||
| filepath = f'detects/{self.model_name}/detects.yml' | ||
|
|
||
| with open(filepath, 'r', encoding='utf-8') as fp: | ||
| self.configs = yaml.safe_load(fp) | ||
|
|
||
| for item in self.configs['templates']: | ||
| if 'template' not in item: | ||
| continue | ||
| item['template'] = cv2.imread(f'detects/{self.model_name}/images/{item["template"]}') | ||
|
|
||
| @staticmethod | ||
| def parse_rect(shape, rect): | ||
| x1 = int(rect['left'] * shape[1]) | ||
| y1 = int(rect['top'] * shape[0]) | ||
| x2 = shape[1] - int(rect['right'] * shape[1]) | ||
| y2 = shape[0] - int(rect['bottom'] * shape[0]) | ||
| return x1, y1, x2, y2 | ||
|
|
||
| def clip_image(self, image, _index): | ||
| index = _index - 1 | ||
| if index >= len(self.configs['templates']): | ||
| logger.warning(f'No such template with index "{_index}"') | ||
| return | ||
| height, width = image.shape[:2] | ||
| template = self.configs['templates'][index] | ||
| x1, y1, x2, y2 = self.parse_rect((height, width), template['rect']) | ||
| filepath = f'detects/{self.model_name}/clips/{template["name"]}.png' | ||
| cv2.imwrite(filepath, image[y1:y2, x1:x2]) | ||
| logger.info(f'Screenshot for "{template["name"]}" saved to {filepath}') | ||
|
|
||
| def mark(self, image, x1, y1, x2, y2, text, color): | ||
| cv2.rectangle(image, (x1, y1), (x2, y2), color, 2) | ||
| if not text: | ||
| return | ||
| rect, _ = cv2.getTextSize(text, self.font, 0.5, 2) | ||
| if y1 < rect[1]: | ||
| cv2.putText(image, text, ((x1 + x2 - rect[0]) // 2, y2 + rect[1]), self. font, 0.5, color, 2) | ||
| else: | ||
| cv2.putText(image, text, ((x1 + x2 - rect[0]) // 2, y1 - rect[1]), self.font, 0.5, color, 2) | ||
|
|
||
| def match_template(self, image): # 模板匹配 | ||
| cover = np.zeros(image.shape, dtype=np.uint8) | ||
| height, width = image.shape[:2] | ||
| groups = {} | ||
|
|
||
| for i, item in enumerate(self.configs['templates']): | ||
| x1, y1, x2, y2 = self.parse_rect((height, width), item['rect']) | ||
| hint = f'[{i+1}] {item["name"]}' | ||
| if 'template' in item: | ||
| target = image[y1:y2, x1:x2] | ||
| template = item['template'] | ||
| self.mark(cover, x1, y1, x2, y2, '', COLOR_YELLOW) | ||
|
|
||
| result = cv2.matchTemplate(target, template, cv2.TM_SQDIFF_NORMED) | ||
| min_val, _, min_loc, _ = cv2.minMaxLoc(result) | ||
| similarity = 1 - min_val # 计算相似度 | ||
| min_loc = min_loc[0] + x1, min_loc[1] + y1 # 还原到绝对坐标 | ||
| min_rect = (*min_loc, min_loc[0] + template.shape[1], min_loc[1] + template.shape[0]) | ||
|
|
||
| hint += f' {similarity * 100:.2f}%' | ||
| if similarity >= item['threshold']: | ||
| color = COLOR_GREEN | ||
| groups[item['name']] = min_rect | ||
| else: | ||
| color = COLOR_RED | ||
|
|
||
| self.mark(cover, *min_rect, hint, color) | ||
| else: | ||
| self.mark(cover, x1, y1, x2, y2, hint, COLOR_BLUE) | ||
|
|
||
| return groups, cover | ||
|
|
||
| def match_progress(self, image, hook_pos, cover): # 进度条匹配 | ||
| progress_config = self.configs['progress'] | ||
|
|
||
| img_height, img_width = image.shape[:2] | ||
| center_x = (hook_pos[0] + hook_pos[2]) // 2 | ||
| center_y = (hook_pos[1] + hook_pos[3]) // 2 - int(img_height * progress_config['offset']) | ||
|
|
||
| width = int(img_width * progress_config['width']) | ||
| height = int(img_height * progress_config['height']) | ||
| x1, x2 = center_x - width // 2, center_x + width // 2 | ||
| y1, y2 = center_y - height // 2, center_y + height // 2 | ||
| progress = image[y1:y2, x1:x2] | ||
|
|
||
| frame_color = progress_config['frame-color'] | ||
| mask = np.all(progress == frame_color, axis=2) | ||
| count = np.sum(mask) | ||
|
|
||
| self.mark(cover, x1, y1, x2, y2, f'Progress [{count}/{width * height}]', (255, 0, 0)) | ||
| if (count / (width * height)) > progress_config['threshold']: | ||
| sample = np.sum(mask, axis=0) | ||
| sp = progress_config['sp'] | ||
| frame_pos, cursor_pos = [], [] | ||
| for i, val in enumerate(sample): | ||
| if sp[0] <= val < sp[1]: | ||
| frame_pos.append(i) | ||
| elif val >= sp[1]: | ||
| cursor_pos.append(i) | ||
| if frame_pos and cursor_pos: | ||
| frame_pos, cursor_pos = [min(frame_pos), max(frame_pos)], [min(cursor_pos), max(cursor_pos)] | ||
| self.mark(cover, x1 + frame_pos[0], y1, x1 + frame_pos[1], y2, 'Frame', (255, 0, 255)) | ||
| cursor_x = sum(cursor_pos) // 2 | ||
| if frame_pos[0] <= cursor_x <= frame_pos[1]: | ||
| color = (0, 255, 0) | ||
| else: | ||
| color = (0, 0, 255) | ||
| self.mark(cover, x1 + cursor_pos[0], y1, x1 + cursor_pos[1], y2, 'Cursor', color) | ||
|
|
||
| # 如果未达到位置则需要点击 | ||
| return cursor_x < frame_pos[0] + (frame_pos[1] - frame_pos[0]) * configs['progress-threshold'] | ||
| return None |
Oops, something went wrong.