def scan(self, image):
print "Saving image..."
output = TemporaryFile.create('.tif')
image.save(output, 'TIFF')
print "Loading image with gamera..."
img = load_image(output)
TemporaryFile.remove(output)
print "Converting to greyscale..."
img = img.to_greyscale()
print "Thresholding..."
onebit = img.otsu_threshold()
# Get connected components from the image
print "Getting connected components"
ccs = onebit.cc_analysis()
# Classify
#classifier = knn.kNNInteractive()
#classifier.from_xml_filename('training.xml')
#classifier.classify_list_automatic( css )
print "Initiating classifier"
classifier = kNNNonInteractive(ccs)
import ocr
o = ocr.Ocr()
print "Scanning with tesseract"
o.scan(image)
print "Teaching gamera"
for c in ccs:
print "Glyph: ", c
text = o.textInRegion(c)
classifier.classify_glyph_manual(c, text)
def ocr_process():
# 载入ocr模型
model = ocr.Ocr(text_process=False)
print("load model done")
while True:
# 尝试从redis数据库取出一队图像数据
queue = db.lrange(settings.IMAGE_QUEUE, 0, settings.BATCH_SIZE - 1)
# 用于放置每幅图像的id
imageIDs = []
batch = None
# 遍历图像数据队列
for q in queue:
# 解码图像数据
q = json.loads(q.decode("utf-8"))
image = helpers.base64_decode_image(q["image"],
settings.IMAGE_DTYPE,
q["shape"])
# 检查数据批是否为空
if batch is None:
batch = image
# 不为空则将数据堆叠起来，形成一批数据
else:
batch = np.vstack([batch, image])
# 更新图像id列表
imageIDs.append(q["id"])
# 检查是否需要进行模型预测
if len(imageIDs) > 0:
# 处理数据块
results = []
print("* Batch size: {}".format(batch.shape))
for img in batch:
preds = model.predict(img)
results.append(preds)
for (imageID, resultSet) in zip(imageIDs, results):
# 初始化预测结果
output = []
# 初始化文本检测结果
boxes = []
for box in resultSet[1]:
boxes.append([box[xy] for xy in range(8)])
r = {"words": resultSet[0], "boxes": boxes}
output.append(r)
# 存储检测与识别结果
db.set(imageID, json.dumps(output, ensure_ascii=False))
db.ltrim(settings.IMAGE_QUEUE, len(imageIDs), -1)
# 延时
time.sleep(settings.SERVER_SLEEP)
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the
# Free Software Foundation, Inc.,
# 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
import ocr
import sys
import codecs
from PySide.QtGui import *
ocr.initOcrSystem()
c = ocr.Ocr()
image = QImage()
if not image.load( sys.argv[-1] ):
print 'Error loading image'
os.exit(1)
c.scan( image )
print c.formatedText().encode('ascii','ignore')
f=codecs.open( '/tmp/output.txt', 'w', 'utf-8' )
f.write( c.formatedText() )
f.close()
print c.slope()
c.deskewOnce()
c.image.save( '/tmp/rotated.png', 'PNG' )
#print "Image stored in /tmp/rotated.png"
def POST(self):
def get_id(img):
_token = gt.get_token()
_url = 'https://api.weixin.qq.com/cgi-bin/media/upload?access_token=' + _token + '&type=image'
print('url=', _url)
payload = {'file': ('upload.jpg', img, 'image/jpg')}
m = MultipartEncoder(payload)
headers = {
'Content-Type': m.content_type,
'other-keys': 'other-values'
}
_id = requests.post(_url, headers=headers, data=m).json()
print('post request returns:', _id)
_id = _id['media_id']
print(_id)
return _id
#try:
webData = web.data()
print("Handle Post webdata is ", webData)
#后台打日志
recMsg = receive.parse_xml(webData)
#print(recMsg.Content.decode('utf-8'))
_data = ''
if isinstance(recMsg, receive.Msg):
#print(urls.triggers)
toUser = recMsg.FromUserName
fromUser = recMsg.ToUserName
if recMsg.MsgType == 'text':
for trigger in urls.triggers:
if trigger in recMsg.Content.decode('utf-8'):
print('got the keyword:', trigger)
for key, item in urls.URL.items():
if item['trigger'] == trigger:
_item = item
break
if _item['type'] == 'txt_event':
print('now making poem...')
_data = _item['keyword']()
print(_data)
replyMsg = reply.TextMsg(toUser, fromUser, _data)
return replyMsg.send()
else:
_data = gf.getData(_item['url'],
data_type=_item['type'])
if _data:
if _item['type'] == 'txt' or _item[
'type'] == 'json':
if _item['type'] == 'json':
_data = _data[_item['keyword']]
replyMsg = reply.TextMsg(
toUser, fromUser, _data)
elif _item['type'] == 'img':
_id = get_id(_data)
replyMsg = reply.ImageMsg(
toUser, fromUser, _id)
return replyMsg.send()
else:
print('request data failed')
_data = ''
break
else:
_data = urls.DEFAULT_MSG
replyMsg = reply.TextMsg(toUser, fromUser, _data)
return replyMsg.send()
elif recMsg.MsgType == 'image':
_url = recMsg.PicUrl
_data = gf.getData(_url, data_type='img')
_data = ocr.Ocr(_data)
_id = get_id(_data)
replyMsg = reply.ImageMsg(toUser, fromUser, _id)
return replyMsg.send()
if _data == '':
print("暂且不处理")
return "success"
def start(self,inputFile, outputFile):
_jsonLib = jsonLib.JsonLib(self._constants)
variables = self._constants._PATH_VARIABLES
textType = _jsonLib.getValue(variables,self._constants._FIELD_TEXT_TYPE)
textTypeCodecs = _jsonLib.getValue(variables,self._constants._FIELD_TEXT_TYPE_CODECS)
languageType = _jsonLib.getValue(variables,self._constants._FIELD_LANGUAGE_CURRENT)
languagePath = _jsonLib.getValue(variables,self._constants._PATH_LANGUAGE)+languageType+self._constants._EXTENSION_JSON
extension = os.path.splitext(inputFile)[self._constants._INDEX_POSITION_EXTENSION][self._constants._INDEX_POSITION_SPLIT_ORDER:]
if (inputFile == self._constants._FIELD_EMPTY_STRING or outputFile == self._constants._FIELD_EMPTY_STRING):
print(_jsonLib.getValue(languagePath,self._constants._FIELD_ERROR_PARAMETERS))
return self._constants._INDEX_RETURN_ERROR
if not (os.path.isfile(inputFile)):
print(_jsonLib.getValue(languagePath,self._constants._FIELD_ERROR_INPUT))
return self._constants._INDEX_RETURN_ERROR
if (not (os.path.dirname(outputFile))) and (os.path.dirname(outputFile) != self._constants._FIELD_EMPTY_STRING):
print(_jsonLib.getValue(languagePath,self._constants._FIELD_ERROR_OUTPUT))
return self._constants._INDEX_RETURN_ERROR
if not (extension in _jsonLib.getValue(variables,self._constants._FIELD_IMAGE_EXTENSION)):
print(_jsonLib.getValue(languagePath,self._constants._FIELD_ERROR_FILE_TYPE))
return self._constants._INDEX_RETURN_ERROR
print(_jsonLib.getValue(languagePath,self._constants._FIELD_MESSAGE_WELCOME))
internalImage = _jsonLib.getValue(variables,self._constants._FIELD_IMAGE_FILENAME)
internalText = _jsonLib.getValue(variables,self._constants._FIELD_TEXT_FILENAME)
_image = image.Image(self._constants)
_image.fixImage(inputFile,internalImage)
if (_image):
del _image
_ocr = ocr.Ocr(self._constants)
_ocr.readPhoto(internalImage,internalText,textTypeCodecs)
if (_ocr):
del _ocr
_spellcheck = spellcheck.Spellcheck(self._constants)
_spellcheck.verifyFile(internalText,languageType,textType,textTypeCodecs)
if (_spellcheck):
del _spellcheck
ttsLanguage = _jsonLib.getValue(languagePath,self._constants._FIELD_GTTS_CODE)
_tts = tts.Tts(self._constants)
audioExists = _tts.convertText(internalText,outputFile,ttsLanguage)
if (_tts):
del _tts
if (audioExists == self._constants._INDEX_RETURN_OK):
print(_jsonLib.getValue(languagePath,self._constants._FIELD_MESSAGE_FINISH))
if (_jsonLib):
del _jsonLib
_functions = functions.Functions(self._constants)
_functions.clearAll(internalImage,internalText)
if (_functions):
del _functions