IMAGE RECOGNITION METHOD
- 總結
- The present invention provides an image recognition method for recognizing a plurality of objects in an image, wherein each object is composed of a plurality of image segments. The image recognition method includes the steps of: sequentially acquiring every pixel of the image; identifying a start point of a newly detected image segment; recording information of the newly detected image segment pixel-by-pixel from the start point; identifying an end point of the newly detected image segment; recognizing an object to which the newly detected image segment belongs according to the start point and the end point of the newly detected image segment; and identifying an invalid object or a merged object thereby releasing the data space thereof.
- 技術優勢
- In the image recognition method of the present invention, when the acquisition of all pixels in one row or one column is accomplished, an object is identified as a complete object when there is no newly detected image segment associated thereto; when a size or an average brightness of a complete object is not within a predetermined range, the complete object is identified as an invalid object; and when the newly detected image segment simultaneously belongs to a plurality of objects, a merged objet is identified.
- 技術應用
- recognition
- 申請日期
- 13-Jan-2011
- 申請號碼
- 13/005,931
- 覆蓋範圍
- display
- 國家/地區
- 美國
欲了解更多信息,請點擊 這裡