From ea3e1870e7f5bc2597327e3cc175de39b02b3239 Mon Sep 17 00:00:00 2001
From: hejianjun <hejianjun11795@foxmail.com>
Date: Fri, 4 Feb 2022 03:49:27 +0800
Subject: [PATCH] =?UTF-8?q?=E5=9B=BE=E7=89=87=E9=9D=A0=E5=8F=B3?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 ImageProcessing/baidu.py | 22 ++++++++++++++++++++++
 ImageProcessing/hog.py   | 15 ++++++++++-----
 WebCrawler/madou.py      | 26 +++++++++++++-------------
 3 files changed, 45 insertions(+), 18 deletions(-)
 create mode 100644 ImageProcessing/baidu.py

diff --git a/ImageProcessing/baidu.py b/ImageProcessing/baidu.py
new file mode 100644
index 0000000..7bbe3ae
--- /dev/null
+++ b/ImageProcessing/baidu.py
@@ -0,0 +1,22 @@
+from aip import AipBodyAnalysis
+import config
+
+
+def face_center(filename, model):
+    app_id = config.getInstance().conf.get("face", "appid")
+    api_key = config.getInstance().conf.get("face", "key")
+    app_secret = config.getInstance().conf.get("face", "secret")
+    client = AipBodyAnalysis(app_id, api_key, app_secret)
+    with open(filename, 'rb') as fp:
+        img = fp.read()
+    result = client.bodyAnalysis(img)
+    if 'error_code' in result:
+        raise result['error_msg']
+    print('[+]Found person      ' + str(result['person_num']))
+    # 中心点取鼻子x坐标
+    max = 0
+    for person_info in result["person_info"]:
+        x = int(person_info['body_parts']['nose']['x'])
+        if x > max:
+            max = x
+    return max
diff --git a/ImageProcessing/hog.py b/ImageProcessing/hog.py
index 28f53fb..6f0bbb3 100644
--- a/ImageProcessing/hog.py
+++ b/ImageProcessing/hog.py
@@ -1,10 +1,15 @@
 import face_recognition
 
+
 def face_center(filename, model):
     image = face_recognition.load_image_file(filename)
-    face_locations = face_recognition.face_locations(image, 0, model)
-    if face_locations:
-        top, right, bottom, left = face_locations[0]
+    face_locations = face_recognition.face_locations(image, 1, model)
+    print('[+]Found person      ' + str(len(face_locations)))
+    max = 0
+    for face_location in face_locations:
+        top, right, bottom, left = face_location
         # 中心点
-        return int((right+left)/2)
-    return 0
\ No newline at end of file
+        x = int((right+left)/2)
+        if x > max:
+            max = x
+    return max
diff --git a/WebCrawler/madou.py b/WebCrawler/madou.py
index 669f6d9..01fc19c 100644
--- a/WebCrawler/madou.py
+++ b/WebCrawler/madou.py
@@ -51,14 +51,6 @@ def getRelease(html):  # 获取出版日期
 def getRuntime(html):  # 获取播放时长
     return ''
 
-
-def getActor(html):  # 获取女优
-    b = []
-    for player in html.xpath('//div[@class="article-tags"]/a/text()'):
-        b.append(player)
-    return b
-
-
 def getUrl(html):
     return str(html.xpath('//a[@class="share-weixin"]/@data-url')[0])
 
@@ -98,11 +90,18 @@ def getTag(html):  # 获取标签
 def getExtrafanart(html):  # 获取剧照
     return ''
 
+def cutTags(tags):
+    actors = []
+    tags = []
+    for tag in tags:
+        actors.append(tag)
+    return actors,tags
+
 
 def main(number):
     try:
         try:
-            number = number.lower()
+            number = number.lower().strip()
             url = "https://madou.club/" + number + ".html"
             htmlcode = get_html(url)
         except:
@@ -110,6 +109,8 @@ def main(number):
 
         html = etree.fromstring(htmlcode, etree.HTMLParser())
         url = getUrl(html)
+        tags = getTag(html)
+        actor,tags = cutTags(tags);
         dic = {
             # 标题
             'title': getTitle(html, number),
@@ -124,7 +125,7 @@ def main(number):
             # 导演
             'director': getDirector(html),
             # 演员
-            'actor': getActor(html),
+            'actor': actor,
             # 发售日
             'release': getRelease(html),
             # 番号
@@ -135,7 +136,7 @@ def main(number):
             'extrafanart': getExtrafanart(html),
             'imagecut': 1,
             #
-            'tag': getTag(html),
+            'tag': tags,
             #
             'label': getSerise(html),
             # 作者图片
@@ -160,5 +161,4 @@ def main(number):
 
 
 if __name__ == '__main__':
-    print(main('MD-0147'))
-    print(main('MD0147'))
+    print(main('MD0094'))