US 12,482,157 B2
Data augmentation device, method, and non-transitory computer readable storage medium
Yung-Hui Li, New Taipei (TW); Van Nhiem Tran, New Taipei (TW); and Chi-En Huang, New Taipei (TW)
Assigned to HON HAI PRECISION INDUSTRY CO., LTD., New Taipei (TW)
Filed by HON HAI PRECISION INDUSTRY CO., LTD., New Taipei (TW)
Filed on Apr. 14, 2023, as Appl. No. 18/300,403.
Claims priority of provisional application 63/362,965, filed on Apr. 14, 2022.
Prior Publication US 2023/0334741 A1, Oct. 19, 2023
Int. Cl. G09G 5/00 (2006.01); G06T 11/60 (2006.01)
CPC G06T 11/60 (2013.01) 14 Claims
OG exemplary drawing
 
1. A data augmentation device, comprising:
a processor, configured to capture at least one sample image from an original image, and to input the at least one sample image to at least two data augmentation modules, so as to generate at least two augmentation image groups through the at least two data augmentation modules;
wherein the at least two data augmentation modules comprise a first data augmentation module and a second data augmentation module, wherein a first parameter group of the first data augmentation module is a fixed value, and a second parameter group of the second data augmentation module is not the fixed value;
wherein the processor is further configured to select at least one pixel point from the original image, to select at least one ratio value between a first value and a second value, and to capture the at least one sample image in accordance to the original image according to the at least one pixel point and the at least one ratio value, wherein the at least one pixel point is a center point of the at least one sample image, the at least one ratio value is a width of the at least one sample image corresponding to a width of the original image, and the at least one ratio value is a height of the at least one sample image corresponding to a height of the original image;
wherein the at least one sample image comprises a first sample image and a second sample image, wherein the processor is further configured to obtain the first sample image according to a first pixel point and the at least one ratio value, and the processor is further configured to obtain the second sample image according to a second pixel point and the at least one ratio value, wherein the first pixel point and the second pixel point are randomly selected;
wherein the processor is further configured to restore the size of the least one sample image to the size of the original image.