| Robot is the offspring of quick developed industry age. In the past years, it was developed very quickly and was widely used in each aspect of human's life and work.Life Assistant Robot was firstly invented for ageing of Japanese population. It is is one type of robots which is an assistant in human's life and work. Contrast to industrial robot such as medical treatment robot and snow clearing robot, it is more close to us. As it is working near human, its working condition is more complicated. Thus how to collect exact information is very important, which is also a study issue.This study work is mainly on image identification via Life Assistant Robot:1. An approach of modeling any space-sheet with any model was developed to solve the limitation of using the only one template. This approach make model divide to two parts: space model and sheet model on HSV Color Space. Using Template Match Method, it calculated similar fitness to identify objects by genetic algorithm. The fittest object is the one to be identified. Then it saved identified result in RF-ID memory and used these saved information to work. 2. A novel modeling approach for identifying any shape which combined HSV Color Space and Template Match Method base on HSV color was also developed. It picked profile of object, got information of profile, filled color in inner of profile and made identification template for identifying any shape. Thus it was more robust. It generated model for identifying for the situation of unsure object and saved every work result in in Radio Frequency-Identification system which was embedded in robots.3. The approach introduced in 2 was applied in Society Circumstance Assistant Robot. This robot got photos from vidicon mounted on its arm and collected information about working conditions via RF-ID system. By using human interface, it got more working information. Then it contrasted photos between templates and used genetic algorithm to calculate similar fitness. Thus it completed identifying, found out the object which human wanted and got desired working effect. When it was used in human's life, it can provide great help for people.4. It studied identification approach of combined HSV Color Space and Canny Operator. It identified profile of color objects, got profile for template using Canny and matched image with templates.All study works make image identification of any shape object via Life Assistant Robot feasible. It can make this approach use widely and help human's life and work. However, a lot of problems need to be studied more. |
PDF Full Text Request