An apparatus and method for composing images in an image processing device are provided. The method includes identifying an object in an overlap region of a plurality of images; generating a Gaussian curve for the object; determining an alpha mask for the overlap region using the Gaussian curve; and
An apparatus and method for composing images in an image processing device are provided. The method includes identifying an object in an overlap region of a plurality of images; generating a Gaussian curve for the object; determining an alpha mask for the overlap region using the Gaussian curve; and blending the plurality of images based on the alpha mask.
대표청구항▼
1. A method for composing images in an image processing device, the method comprising: identifying an object in an overlap region of a plurality of images;generating a Gaussian curve for the object;determining an alpha mask for the overlap region using the Gaussian curve; andblending the plurality o
1. A method for composing images in an image processing device, the method comprising: identifying an object in an overlap region of a plurality of images;generating a Gaussian curve for the object;determining an alpha mask for the overlap region using the Gaussian curve; andblending the plurality of images based on the alpha mask. 2. The method of claim 1, wherein the Gaussian curve is generated using: y=aⅇ-(x-b)22c2,where a represents a peak/maximum value of the Gaussian curve, b represents a mean value of the Gaussian curve, e represents Euler's number, and c represents a width of the Gaussian curve, based on: c=HWHM22ln2where HWHM is a Half Width at Half Maximum. 3. The method of claim 2, wherein the Gaussian curve is determined by: a=hb=xoffset+wobject2c=2wobject22ln2,where h represents a height of the object in the overlap region, xoffset represents a width from a start of the overlap region to the object in the overlap region, and wobject represents a width of the object in the overlap region. 4. The method of claim 2, wherein when two objects are identified in the overlap region, generating the Gaussian curve, further comprises super-positioning individual Gaussian curves centered around the objects. 5. The method of claim 2, wherein when two objects are identified in the overlap region, generating the Gaussian curve further comprises identifying a minimum value of overlapping curves of the objects. 6. The method of claim 1, wherein the alpha mask is determined using a gradient equation given by: α=αmax×(rowd)k,where k is a positive real number, row=0 at top of the overlap region and increases downwards, d represents a distance from row=0 to the Gaussian curve, and αmax is a maximum value of the alpha mask. 7. An apparatus for composing images in an image processing device, the apparatus comprising: a display unit; andan image blending device for identifying an object in an overlap region of a plurality of images, for generating a Gaussian curve for the object, for determining an alpha mask for the overlap region using the Gaussian curve, and for blending the plurality of images based on the alpha mask. 8. The apparatus of claim 7, wherein the Gaussian curve is generated using: y=aⅇ-(x-b)22c2,where a represents a peak/maximum value of the Gaussian curve, b represents a mean value of the Gaussian curve, e represents Euler's number, and c represents a width of the Gaussian curve, based on: c=HWHM22ln2where HWHM is a Half Width at Half Maximum. 9. The apparatus of claim 7, wherein the Gaussian curve is determined by: a=hb=xoffset+wobject2c=2wobject22ln2where h represents a height of the object in the overlap region, xoffset represents a width from a start of the overlap region to the object in the overlap region, and wobject represents a width of the object in the overlap region. 10. The apparatus of claim 7, wherein when two objects are identified in the overlap region, the image blending device generates the Gaussian curve by super-positioning of individual Gaussian curves centered around the objects. 11. The apparatus of claim 7, wherein when two objects are identified in the overlap region, the image blending device generates the Gaussian curve by using a minimum value of overlapping curves of the objects. 12. The apparatus of claim 7, wherein the alpha mask is determined using a gradient equation given by: α=αmax×(rowd)k,where k is a positive real number, row=0 at top of the overlap region and increases downwards, d represents a distance from row=0 to the Gaussian curve, and αmax is a maximum value of the alpha mask.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (3)
Yalin Xiong ; Ken Turkowski, Aligning rectilinear images in 3D through projective registration and calibration.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.