闲来无事，写个程序编排财大课表，也算是催促自己学习
项目地址：https://github.com/PaperMisty/Tjufe-course-arrange
其中，denoise2.py程序主要将课表src.png降噪为tmp.png，再识别输出到info.txt，由于课表本身太糙，噪点也多，识别起来难免有错误，于是需要对比着自己改一下；
确保无误后保存，info2ics.py程序将info.txt编排为两份ics日历文件，一个是GMT.ics（格林威治时间），一个是Beijing.ics（北京时间），取自己需要的导入即可。
如有错误，欢迎交流。
1 # encoding:utf-8
2 import cv2
3 import numpy as np
4 import ocr
5 # 严苛的去噪版本
6
7 # 读入图像并转为灰度图像
8 src = cv2.imread('src.png')
9 img = cv2.cvtColor(src,cv2.COLOR_BGR2GRAY)
10
11
12 # 定义去噪函数
13 def img_denoise(img,kernel,noise_num):
14 output = img.copy()
15 # 获取图像大小
16 height, width = img.shape[:2]
17 # 循环遍历图像每一个像素
18 for i in range(1, height - 1):
19 for j in range(1, width - 1):
20 # 获取kernel x kernel核的像素值
21 roi = img[i - 1:i + kernel - 1, j - 1:j + kernel - 1]
22 # 判断是否有不少于noise_num个像素近于0，判定为黑色成行或成列，进行保留；否则为孤点，取值255为白
23 if np.sum(roi < 130) >= noise_num:
24 output[i, j] = img[i, j]
25 else:
26 output[i, j] = 255
27 return output
28 img = img_denoise(img,3,2)
29 # 二次过滤
30 img = img_denoise(img,5,3)
31
32 # 显示输出图像
33 # cv2.imshow('img', img)
34 # cv2.waitKey(0)
35 # cv2.destroyAllWindows()
36
37 # 去噪图片储存
38 tmp_png_name = 'tmp.png'
39 cv2.imwrite(tmp_png_name,img)
40 # 获取识别结果文本
41 response = ocr.ocr_course(tmp_png_name)
42 # 文本整理输入到info.txt
43 ocr.info_format(response)
denoise2.py
1 # encoding:utf-8
2 import requests
3 import base64
4 import pandas as pd
5
6 # 利用百度识别api识别图片文本信息
7 def ocr_course(path):
8 # client_id 为官网获取的AK， client_secret 为官网获取的SK
9 host = 'https://aip.baidubce.com/oauth/2.0/token?grant_type=client_credentials&client_id=CyPGn6Z0tCOxG747tru6KRb9&client_secret=2Fr6y4W5Gezp6H2RHt6OVx6uEXUuOOUz'
10 tokens = requests.get(host).json()
11
12 request_url = "https://aip.baidubce.com/rest/2.0/ocr/v1/accurate_basic"
13 # 二进制方式打开图片文件
14 f = open(path, 'rb')
15 img = base64.b64encode(f.read())
16 # 定义申请识别的参数
17 params = {"image":img}
18 access_token = tokens['access_token']
19 request_url = request_url + "?access_token=" + access_token
20 headers = {'content-type': 'application/x-www-form-urlencoded'}
21 # 获取识别结果
22 response = requests.post(request_url, data=params, headers=headers)
23 return response
24
25 # 自定义文本纠正
26 def week_fix(str):
27 if str[0] == '-' or str[0] == '~':
28 str = '一'+str[1:]
29 if str[-1] == '欢':
30 str = str[:-1]+'双'
31 return str
32
33 # 将识别结果输入到info.txt
34 def info_format(response):
35 if response:
36 seq = ''
37 information = ''
38 for item in response.json()['words_result']:
39 information += item['words']+' '+seq
40 # 将结果整理分开
41 if seq == '\n':
42 seq = ''
43 if "节" in item['words']:
44 seq = '\n'
45 else:
46 return None
47 #df = pd.read_excel('info.xlsx','w',sheet_name='Sheet1')
48 with open('info.txt','w',encoding='utf-8') as f:
49 f.write("#示例(除了XXX其余都需要仔细校对)：")
50 f.write("XXXXXXXXX]课程名字 老师名字 1-17 二[5XXX]双 N3218"+'\n')
51 content = ''
52 tmp = ''
53 for i in information:
54 if i != '\n':
55 # 把每行内容累积起来
56 content += i
57 else:
58 # 换行，写入文本文件
59 info_list = content.split(' ')
60 # 自定义纠正
61 info_list[-3] = week_fix(info_list[-3])
62 # 如果这条是上一课程信息的续写，
63 if len(info_list) < 6:
64 f.write(tmp+' '+info_list[-4]+' '+info_list[-3]+' '+info_list[-2]+' '+'\n')
65 # 如果这是新的课程
66 else:
67 f.write(info_list[1]+' '+info_list[-5]+' '+info_list[-4]+' '+info_list[-3]+' '+info_list[-2]+' '+'\n')
68 # 将这次部分内容暂存，供下次可能的使用
69 tmp = info_list[1]+' '+info_list[-5]
70
71 content = ''
72 #info_format(response)
ocr.py
1 # 北京时间：GMT+8:00
2 import uuid
3 from datetime import datetime, timedelta
4
5 def create_ics_file(events,calendar_name):
6 # 生成ics文件头部信息
7 ics_file = 'BEGIN:VCALENDAR\nVERSION:2.0\nPRODID:-//hacksw/handcal//NONSGML v1.0//EN\n'
8
9 # 循环生成每个事件的ics信息
10 for event in events:
11 # 生成事件唯一标识符
12 event_uid = uuid.uuid4().hex
13
14 # 生成事件开始和结束时间
15 start_time = event['start_time']
16 end_time = event['end_time']
17
18 # 生成事件的ics信息
19 ics_event = f'BEGIN:VEVENT\nUID:{event_uid}\nDTSTAMP:{datetime.now().strftime("%Y%m%dT%H%M%SZ")}\nDTSTART:{start_time.strftime("%Y%m%dT%H%M%S")}\nDTEND:{end_time.strftime("%Y%m%dT%H%M%S")}\nSUMMARY:{event["summary"]}\nDESCRIPTION:{event["description"]}\nLOCATION:{event["location"]}\nEND:VEVENT\n'
20
21 # 将事件信息添加到ics文件中
22 ics_file += ics_event
23
24 # 添加ics文件尾部信息
25 ics_file += 'END:VCALENDAR\n'
26
27 # 将ics文件保存到本地
28 with open(calendar_name, 'w', encoding='utf-8') as f:
29 f.write(ics_file)
30
31 path = 'info.txt'
32 with open(path,'r',encoding='utf-8') as f:
33 contents = f.readlines()
34 # 从非注释行开始
35 structs = []
36 for content in contents[1:]:
37 sub_dict = {}
38 con_list = content.split(' ')
39 # 课程
40 sub_dict['summary'] = con_list[0].split(']')[1]
41 # 老师
42 sub_dict['description'] = con_list[1]
43 # 周数
44 weeks = con_list[2]
45 if '-' in weeks:
46 sub_dict['week_start'] = int(weeks.split('-')[0])
47 sub_dict['week_end'] = int(weeks.split('-')[1])
48 else:
49 sub_dict['week_start'] = int(weeks)
50 sub_dict['week_end'] = int(weeks)
51 # 日程
52 schedule = con_list[3]
53 sub_dict['week_time'] = schedule[0]
54 # 默认只上两节（一单）课
55 schedule = schedule[2:]
56 sub_dict['start_course'] = schedule.split('-')[0]
57 if schedule[-1] == "单":
58 sub_dict['step'] = 2
59 sub_dict['week_parity'] = 0
60 elif schedule[-1] == "双":
61 sub_dict['step'] = 2
62 sub_dict['week_parity'] = 1
63 else:
64 sub_dict['step'] = 1
65 sub_dict['week_parity'] = 0
66 # 位置
67 sub_dict['location'] = con_list[4]
68
69 structs.append(sub_dict)
70
71 def arrange(time_block):
72 events = []
73 # 课程遍历
74 for struct in structs:
75 # 周数遍历
76 days = -1
77 for week in range(struct['week_start']+struct['week_parity'],struct['week_end']+1,struct['step']):
78 sub_dict2 = {}
79 # 周的累计
80 days += 7*(week-1)
81 # 周具体时间修改
82 if struct['week_time'] == '一':
83 days += 1
84 elif struct['week_time'] == '二':
85 days += 2
86 elif struct['week_time'] == '三':
87 days += 3
88 elif struct['week_time'] == '四':
89 days += 4
90 elif struct['week_time'] == '五':
91 days += 5
92 elif struct['week_time'] == '六':
93 days += 6
94 elif struct['week_time'] == '日':
95 days += 7
96 # 教学楼时间修改
97 if struct['location'][0] in 'CDEFJM':
98 # 时段修改
99 if struct['start_course'] == '1':
100 hours = 0
101 minutes = 0
102 elif struct['start_course'] == '3':
103 hours = 2
104 minutes = 10
105 elif struct['start_course'] == '5':
106 hours = 5
107 minutes = 20
108 elif struct['start_course'] == '7':
109 hours = 7
110 minutes = 30
111 elif struct['start_course'] == '9':
112 hours = 9
113 minutes = 50
114 elif struct['start_course'] == '11':
115 hours = 11
116 minutes = 40
117 sub_dict2['start_time'] = datetime(2023, 2, 27, time_block, 10, 0)+timedelta(days=days,minutes=minutes,hours=hours)
118 sub_dict2['end_time'] = sub_dict2['start_time']+timedelta(minutes=30,hours=1)
119 sub_dict2['summary'] = struct['summary']
120 sub_dict2['description'] = struct['description']
121 sub_dict2['location'] = struct['location']
122 else:
123 if struct['start_course'] == '1':
124 hours = 0
125 minutes = 0
126 elif struct['start_course'] == '3':
127 hours = 2
128 minutes = 0
129 elif struct['start_course'] == '5':
130 hours = 5
131 minutes = 20
132 elif struct['start_course'] == '7':
133 hours = 7
134 minutes = 20
135 elif struct['start_course'] == '9':
136 hours = 10
137 minutes = 0
138 elif struct['start_course'] == '11':
139 hours = 11
140 minutes = 50
141 sub_dict2['start_time'] = datetime(2023, 2, 27, time_block, 0, 0)+timedelta(days=days,minutes=minutes,hours=hours)
142 sub_dict2['end_time'] = sub_dict2['start_time']+timedelta(minutes=30,hours=1)
143 sub_dict2['summary'] = struct['summary']
144 sub_dict2['description'] = struct['description']
145 sub_dict2['location'] = struct['location']
146 events.append(sub_dict2)
147 # 结束一周，days归位
148 days = -1
149 print(f'课程总数：{len(events)}')
150 return events
151 GMT_events = arrange(8)
152 Beijing_events = arrange(0)
153 create_ics_file(GMT_events,"GMT.ics")
154 create_ics_file(Beijing_events,"Beijing.ics")
info2ics.py
src.png图片示例：
标签：info,week,财大,sub,课表,start,编排,time,struct From： https://www.cnblogs.com/PaperMisty/p/17151978.html