{"id":2774,"date":"2019-06-02T11:31:33","date_gmt":"2019-06-02T02:31:33","guid":{"rendered":"http:\/\/bezelie.com\/flitz\/?page_id=2774"},"modified":"2020-02-11T17:29:54","modified_gmt":"2020-02-11T08:29:54","slug":"%e3%82%ab%e3%83%a1%e3%83%a9%e3%81%a7%e9%a1%94%e8%aa%8d%e8%ad%98","status":"publish","type":"page","link":"http:\/\/bezelie.com\/flitz\/%e3%82%ab%e3%83%a1%e3%83%a9%e3%81%a7%e9%a1%94%e8%aa%8d%e8%ad%98\/","title":{"rendered":"\u30ab\u30e1\u30e9\u3067\u9854\u8a8d\u8b58"},"content":{"rendered":"\n

\u30fb\u3053\u3053\u3067\u306f\u3001\u3079\u30bc\u30ea\u30fc\u306b\u5185\u8535\u3057\u305f\u30ab\u30e1\u30e9\u3067\u4eba\u9593\u306e\u9854\u3092\u8a8d\u8b58\u3055\u305b\u3066\u307f\u307e\u3059\u3002<\/p>\n\n\n\n

\u30fb\u9854\u306e\u8a8d\u8b58\u306b\u306f\u30aa\u30fc\u30d7\u30f3\u30bd\u30fc\u30b9\u306e\u753b\u50cf\u51e6\u7406\u30e9\u30a4\u30d6\u30e9\u30ea\u300cOpen CV\uff08Open Source Computer Vision Library\uff09\u300d\u3092\u4f7f\u3044\u307e\u3059\u3002<\/p>\n\n\n\n

\u6e96\u5099<\/h5>\n\n\n\n

\u30fb\u307e\u305a\u306fPython3\u306bOpen CV\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b\u5fc5\u8981\u304c\u3042\u308a\u307e\u3059\u3002python\u3084Raspbian\u306e\u30d0\u30fc\u30b8\u30e7\u30f3\u306b\u3088\u3063\u3066\u306f\u5de7\u304f\u3044\u304b\u306a\u3044\u304b\u3082\u3057\u308c\u307e\u305b\u3093\u3002\u4e0d\u5e78\u306b\u3082\u30a8\u30e9\u30fc\u304c\u3067\u3066\u3057\u307e\u3063\u305f\u5834\u5408\u306f\u3001\u6e08\u307f\u307e\u305b\u3093\u304c\u5f53\u65b9\u3067\u306f\u5bfe\u5fdc\u3057\u304d\u308c\u307e\u305b\u3093\u306e\u3067\u3001\u9811\u5f35\u3063\u3066\u30b0\u30b0\u3063\u3066\u307f\u3066\u304f\u3060\u3055\u3044\u3002\u306a\u304a\u3001\u3042\u304f\u307e\u3067Python2\u3067\u306f\u306a\u304fPython3\u306b\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3057\u306a\u3051\u308c\u3070\u306a\u3089\u306a\u3044\u306e\u3067\u6ce8\u610f\u304c\u5fc5\u8981\u3067\u3059\u3002<\/p>\n\n\n\n

\u30fb\u7b46\u8005\u306e\u74b0\u5883\u306e\u5834\u5408\u306f\u4ee5\u4e0b\u306e\u30b3\u30de\u30f3\u30c9\u3067\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u304c\u3067\u304d\u3066\u304a\u308a\u307e\u3059\u304c\u3001\u6050\u7e2e\u306a\u304c\u3089\u52d5\u4f5c\u4fdd\u8a3c\u306f\u3067\u304d\u307e\u305b\u3093\u3002<\/p>\n\n\n\n

$ sudo apt-get install libhdf5-dev libhdf5-serial-dev libhdf5-100
$ sudo apt-get install libqtgui4 libqtwebkit4 libqt4-test python3-pyqt5
$ sudo apt-get install libatlas-base-dev
$ sudo apt-get install libjasper-dev
$ sudo pip3 --default-timeout=1000 install opencv-python
<\/pre>\n\n\n\n
\u30fb\u300c\u30e9\u30ba\u30d1\u30a4\u30ab\u30e1\u30e9<\/a>\u300d\u3092\u30af\u30ea\u30a2\u3057\u305f\u72b6\u614b\uff08\u30e9\u30ba\u30d1\u30a4\u30ab\u30e1\u30e9\u3092\u63a5\u7d9a\u3057\u3001HDMI\u30c7\u30a3\u30b9\u30d7\u30ec\u30a4\u304c\u63a5\u7d9a\u3055\u308c\u305f\u72b6\u614b\uff09\u306b\u3057\u3066\u304a\u3044\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n\n\n\n
\u30b5\u30f3\u30d7\u30eb\u30d7\u30ed\u30b0\u30e9\u30e0\u3000sample_face1.py<\/h5>\n\n\n\n
\u30fb\u30d1\u30bd\u30b3\u30f3\u304b\u3089\u306e\u9060\u9694\u64cd\u4f5c\u3060\u3068\u52d5\u753b\u304c\u8868\u793a\u3055\u308c\u306a\u3044\u3053\u3068\u304c\u3042\u308b\u306e\u3067\u3001\u30e9\u30ba\u30d1\u30a4\u306b\u76f4\u63a5\u3064\u306a\u3052\u305f\u30ad\u30fc\u30dc\u30fc\u30c9\u3068\u30c7\u30a3\u30b9\u30d7\u30ec\u30a4\u3067\u5b9f\u884c\u3057\u3066\u304f\u3060\u3055\u3044\u3002\uff08\u56de\u907f\u624b\u6bb5\u306f\u3042\u308a\u307e\u3059\u304c\u3001\u3053\u3053\u3067\u306f\u8aac\u660e\u3057\u307e\u305b\u3093\uff09<\/p>\n\n\n\n
$ python3 sample_face1.py<\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\u30fb\u30c7\u30a3\u30b9\u30d7\u30ec\u30a4\u306b\u8868\u793a\u3055\u308c\u3066\u3044\u308b\u30ab\u30e1\u30e9\u753b\u50cf\u306e\u4e2d\u306b\u9854\uff08\u306e\u3088\u3046\u306a\u3082\u306e\uff09\u304c\u53ce\u307e\u308b\u3068\u3001\u8d64\u3044\u56db\u89d2\u304c\u8868\u793a\u3055\u308c\u3001\u30c7\u30a3\u30b9\u30d7\u30ec\u30a4\u306b\u300c\u9854\u3092\u767a\u898b\u300d\u3068\u8868\u793a\u3055\u308c\u308b\u306f\u305a\u3067\u3059\u3002\u9854\u304c\u753b\u9762\u304b\u3089\u306f\u307f\u51fa\u3066\u3044\u305f\u308a\u3001\u50be\u3044\u3066\u3044\u305f\u308a\u3001\u9060\u3059\u304e\u308b\u3068\u53cd\u5fdc\u3057\u307e\u305b\u3093\u3002
\u30fb\u7d42\u4e86\u3057\u305f\u3044\u3068\u304d\u306f\u30ad\u30fc\u30dc\u30fc\u30c9\u306e\u300cq\u300d\u30dc\u30bf\u30f3\u3092\u62bc\u3057\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n\n\n\n
\u30bd\u30fc\u30b9\u30b3\u30fc\u30c9<\/h5>\n\n\n\n
#!\/usr\/bin\/env python3\n# -- coding: utf-8 --\n# Bezelie Sample Code : Face Recognition Test\n import picamera                        # \u30ab\u30e1\u30e9\u7528\u30e2\u30b8\u30e5\u30fc\u30eb\n import picamera.array                  # \u30ab\u30e1\u30e9\u7528\u30e2\u30b8\u30e5\u30fc\u30eb\n import cv2                             # Open CV\u30e2\u30b8\u30e5\u30fc\u30eb    \n import bezelie                         # \u3079\u30bc\u30ea\u30fc\u5c02\u7528\u30b5\u30fc\u30dc\u5236\u5fa1\u30e2\u30b8\u30e5\u30fc\u30eb\n\n cascade_path =  \"haarcascade_frontalface_alt.xml\" # \u9854\u8a8d\u8b58xml\n cascade = cv2.CascadeClassifier(cascade_path)\n\n# \u30b5\u30fc\u30dc\u306e\u6e96\u5099\n bez = bezelie.Control()               # \u3079\u30bc\u30ea\u30fc\u64cd\u4f5c\u30a4\u30f3\u30b9\u30bf\u30f3\u30b9\u306e\u751f\u6210\n bez.moveCenter()                      # \u30b5\u30fc\u30dc\u3092\u30bb\u30f3\u30bf\u30ea\u30f3\u30b0\n \n#\u30e1\u30a4\u30f3\u30eb\u30fc\u30d7\n def main():\n   with picamera.PiCamera() as camera:                         # Open Pi-Camera as camera\n     with picamera.array.PiRGBArray(camera) as stream:         # Open Video Stream from Pi-Camera as stream\n       camera.resolution = (640, 480)                          # Display Resolution\n       camera.hflip = True                                     # Vertical Flip \n       camera.vflip = True                                     # Horizontal Flip\n       while True:\n         camera.capture(stream, 'bgr', use_video_port=True)    # Capture the Video Stream\n         gray = cv2.cvtColor(stream.array, cv2.COLOR_BGR2GRAY) # Convert BGR to Grayscale\n         facerect = cascade.detectMultiScale(gray,             # Find face from gray\n           scaleFactor=1.9,                                    # 1.1 - 1.9 :the bigger the quicker & less acurate \n           minNeighbors=3,                                     # 3 - 6 : the smaller the more easy to detect\n           minSize=(80,120),                                  # Minimam face size \n           maxSize=(640,480))                                  # Maximam face size\n         if len(facerect) > 0:\n           for rect in facerect:\n             cv2.rectangle(stream.array,                       # Draw a red rectangle at face place \n               tuple(rect[0:2]),                               # Upper Left\n               tuple(rect[0:2]+rect[2:4]),                     # Lower Right\n               (0,0,255), thickness=2)                         # Color and thickness\n           print (\"\u9854\u3092\u767a\u898b\")\n         cv2.imshow('frame', stream.array)                     # Display the stream\n         bez.movePitch (1, 0)\n         if cv2.waitKey(1) & 0xFF == ord('q'):                 # Quit operation\n           break\n         stream.seek(0)                                        # Reset the stream\n         stream.truncate()\n       cv2.destroyAllWindows()\n \nif name == \"main\":\n   main()<\/pre>\n\n\n\n
\u89e3\u8aac<\/h5>\n\n\n\n
\u30fb\uff19~\uff11\uff10\u884c\u76ee\uff1a\u9854\u8a8d\u8b58\u306e\u305f\u3081\u306e\u5206\u985e\u5668\u3092\u8aad\u307f\u8fbc\u3093\u3067\u3044\u307e\u3059\u3002
\u30fb\uff12\uff14\u884c\u76ee\uff1a\u30ab\u30e1\u30e9\u306e\u753b\u50cf\u3092RGB\u3067\u30ad\u30e3\u30d7\u30c1\u30e3\u30fc\u3057\u3066\u3044\u307e\u3059\u3002
\u30fb\uff12\uff15\u884c\u76ee\uff1aRGB\u52d5\u753b\u3092\u30b0\u30ec\u30fc\u30b9\u30b1\u30fc\u30eb\u306b\u5909\u63db\u3057\u3066\u3044\u307e\u3059\u3002
\u30fb\uff12\uff16~\uff13\uff10\u884c\u76ee\uff1a\u9854\u3092\u8a8d\u8b58\u3057\u3066\u3044\u307e\u3059\u3002
\u30fb\uff13\uff11~\uff13\uff17\u884c\u76ee\uff1a\u9854\u304c\u898b\u3064\u304b\u3063\u305f\u5834\u5408\u306e\u51e6\u7406\u3067\u3059\u3002<\/p>\n\n\n\n
\u5fdc\u7528<\/h5>\n\n\n\n
\u30fb\uff12\uff17~\uff13\uff10\u884c\u76ee\u304c\u9854\u3092\u8a8d\u8b58\u3059\u308b\u305f\u3081\u306e\u30d1\u30e9\u30e1\u30fc\u30bf\u3067\u3059\u3002\u3044\u308d\u3044\u308d\u6570\u5024\u3092\u3044\u3058\u3063\u3066\u307f\u307e\u3057\u3087\u3046\u3002\u4f8b\u3048\u3070minSize\u306f\u9854\u3068\u3057\u3066\u8a8d\u8b58\u3059\u308b\u6700\u4f4e\u306e\u30b5\u30a4\u30ba\u3067\u3059\u306e\u3067\u3001\u3053\u308c\u3092\u5c0f\u3055\u304f\u3059\u308b\u3068\u3001\u9060\u304f\u306e\u9854\u3082\u8a8d\u8b58\u3055\u308c\u308b\u3088\u3046\u306b\u306a\u308a\u307e\u3059\u3002\u305d\u306e\u304b\u308f\u308a\u3001\u58c1\u306b\u8cbc\u3063\u3066\u3042\u308b\u5199\u771f\u306a\u3069\u3001\u9854\u3058\u3083\u306a\u3044\u3082\u306e\u3092\u8aa4\u8a8d\u8b58\u3059\u308b\u78ba\u7387\u304c\u4e0a\u304c\u3063\u3066\u3057\u307e\u3046\u304b\u3082\u3057\u308c\u307e\u305b\u3093\u3002<\/p>\n\n\n\n
\n\n\n\n
\u5fdc\u7528\u30d7\u30ed\u30b0\u30e9\u30e0 demo_face1.py<\/h5>\n\n\n\n
\u9854\u8a8d\u8b58\u30d7\u30ed\u30b0\u30e9\u30e0\u3092\u6539\u9020\u3057\u305f\u3082\u306e\u304c\u4ee5\u4e0b\u306e\u30d7\u30ed\u30b0\u30e9\u30e0\u306b\u306a\u308a\u307e\u3059\u3002\u30e9\u30f3\u30c0\u30e0\u3067\u5411\u304d\u3092\u5909\u3048\u3001\u9854\u3092\u8a8d\u8b58\u3057\u305f\u3089\u558b\u308a\u307e\u3059\u3002\u30bb\u30ea\u30d5\u30ea\u30b9\u30c8\u300cdata_faceDialogJ.csv\u300d\u3092\u66f8\u304d\u63db\u3048\u308b\u3053\u3068\u3067\u30bb\u30ea\u30d5\u3092\u5909\u3048\u308b\u3053\u3068\u3082\u3067\u304d\u307e\u3059\u306e\u3067\u3001\u5e97\u982d\u3084\u30d6\u30fc\u30b9\u3084\u53d7\u4ed8\u306b\u7f6e\u3044\u305f\u308a\u3059\u308b\u3068\u3001\u5ba2\u5bc4\u305b\u306b\u306a\u308b\u306e\u3067\u306f\u306a\u3044\u304b\u3068\u601d\u3044\u307e\u3059\u3002<\/p>\n\n\n\n
python3 demo_face1.py<\/pre>\n\n\n\n
\n