2012 Fiscal Year Final Research Report
Quantitative evaluation of photographic diet record using computer vision
Project/Area Number |
22650185
|
Research Category |
Grant-in-Aid for Challenging Exploratory Research
|
Allocation Type | Single-year Grants |
Research Field |
Eating habits, studies on eating habits
|
Research Institution | Chukyo University |
Principal Investigator |
|
Co-Investigator(Kenkyū-buntansha) |
OIDA Yukio 中京大学, 情報理工学部, 教授 (30185178)
IMAEDA Nahomi 名古屋女子大学, 家政学部, 准教授 (80387662)
|
Project Period (FY) |
2010 – 2012
|
Keywords | コンピュータビジョン / 食事調査 |
Research Abstract |
It became mandatory for us to the specific health checkup and the specific counseling guidance. There are many people who need to improve his dietary in preventing lifestyle-related diseases. The photographic dietary survey method and 24 hours recall method are often done as the dietary counseling. In the case of recall based method, ambiguity and forgetting are problem. In the case of the photographic dietary survey method, it is difficult to measure the accurate volume of food, because there are many disturbances such as occlusions, varieties of dishes. In this paper, we propose a system which measures the size of the food by stereoscopic camera for assisting the nutrition manager. In the experiments, our system could measure the several kinds of hamburger. It is known that the accuracy of estimated food volume is improved by using our system.
|
Research Products
(4 results)