Method for converting two-dimensional images into three-dimensional images

The invention discloses a method for converting two-dimensional images into three-dimensional images, which comprises steps of firstly, receiving a two-dimensional image represented by red, green and blue; secondly, converting the two-dimensional image into the YUV (luma and chroma) color space and extracting a luminance signal space image; thirdly, dividing the luminance signal space image into multiple sub-images and performing wavelet analysis on each sub-image to obtain wavelet coefficient and average high frequency coefficient of each sub-image; fourthly, obtaining seed nodes for region growing to obtain positions and outlines of objects; fifthly, processing borders of the objects in the image according to the sizes of the obtained outlines; sixthly, smoothly filtering the luminance signal space image to generate a depth image; and seventhly, using the depth image drawing principle combined with the depth image and the two-dimensional image to generated the three-dimensional image. The two-dimensional images can be converted into the three-dimensional images by using the method without other information or extra image capturing equipment.