Image gray stereo conversion method based on GSCGAN

• Main Authors: BYUN EUN-WOOK, SHEN HUA, JIAO KERONG, XING TAO, ZHU CHANGMEI, SHI QIANYOU
• Format: Patent
• Language: Chinese; English
• Published: 10.06.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention discloses an image grayscale stereo conversion method based on GSCGAN. The method comprises the following steps: step 1, constructing a bright field mode image set, a phase contrast mode image set and a differential interference mode image set; 2, a light propagation convolution formula in phase contrast imaging and differential interference imaging is used for replacing convolution propagation operation in a traditional GAN to construct a complex value GAN, and a preliminary GSCGAN model is established; step 3, inputting the training image into the preliminary GSCGAN model for training to obtain a GSCGAN model; and 4, inputting the transparent target sample image into the GSCGAN model for testing to obtain a result image. According to the invention, a model based on GSCGAN is adopted to carry out gray scale stereo conversion on a transparent sample image, and an image with a phase contrast interference effect or a differential interference effect is obtained. The constructed mode discriminator