您的位置: 首页 > 农业专利 > 详情页

Assessment of the neural network for eye image segmentation and picture quality
专利权人:
MAGIC LEAP; INC.
发明人:
SPIZHEVOY, ALEXEY,KAEHLER, ADRIAN,BADRINARAYANAN, VIJAY
申请号:
CA3038031
公开号:
CA3038031A1
申请日:
2017.05.25
申请国别(地区):
CA
年份:
2018
代理人:
摘要:
The system and method for eye image segmentation disclose the estimation of quality. In one aspect, on receiving eye image, equipment such as augmented reality equipment can handle eye image using neural network and the combined structure of convolution to generate on two segmentation eye images and one estimates eye image. Segmentation eye image may include background area, sclera region, iris region, pupil area. On the other hand, the neural network of convolution can be trained for estimating eye image segmentation and picture quality with combined structure. In yet another aspect, segmentation left-eye image can be used to eye contour is determined, for example, the profile of pupil and iris profile in equipment. The present apparatus can make the calculating iris code or biometric authentication of eye contour creation polar region image iris region.
来源网站:
中国工程科技知识中心
来源网址:
http://www.ckcest.cn/home/

意 见 箱

匿名:登录

个人用户登录

找回密码

第三方账号登录

忘记密码

个人用户注册

必须为有效邮箱
6~16位数字与字母组合
6~16位数字与字母组合
请输入正确的手机号码

信息补充