Abstract: Disclosed herein is attack-less adversarial training for robust adversarial defense. The attack-less adversarial training for robust adversarial defense includes the steps of: (a) generating individual intervals (ci) by setting the range of color (C) and then discretizing the range of color (C) by a predetermined number (k); (b) generating one batch from an original image (X) and training a learning model with the batch; (c) predicting individual interval indices (?ialat) from respective pixels (xi) of the original image (X) by using an activation function; (d) generating a new image (Xalat) through mapping and randomization; and (e) training a convolutional neural network with the image (Xalat) generated in step (d) and outputting a predicted label (?).

Abstract: Disclosed herein is attack-less adversarial training for robust adversarial defense. The attack-less adversarial training for robust adversarial defense includes the steps of: (a) generating individual intervals (ci) by setting the range of color (C) and then discretizing the range of color (C) by a predetermined number (k); (b) generating one batch from an original image (X) and training a learning model with the batch; (c) predicting individual interval indices (?ialat) from respective pixels (xi) of the original image (X) by using an activation function; (d) generating a new image (Xalat) through mapping and randomization; and (e) training a convolutional neural network with the image (Xalat) generated in step (d) and outputting a predicted label (?).

Abstract: Disclosed is a depth-priority integral imaging display method using a nonuniform dynamic mask array which makes it possible to enhance the resolutions in both horizontal and vertical directions in such a way to change a division boundary of a mask pattern and use a two-times time multiplexing. It is possible to enjoy the images of clear resolutions even in the product of 60 Hz, and the afterimages at the division boundary may be eliminated by changing the dividing direction of the mask, so the images with clearer resolutions can be displayed.

Abstract: Disclosed is a depth-priority integral imaging display method using a nonuniform dynamic mask array which makes it possible to enhance the resolutions in both horizontal and vertical directions in such a way to change a division boundary of a mask pattern and use a two-times time multiplexing. It is possible to enjoy the images of clear resolutions even in the product of 60 Hz, and the afterimages at the division boundary may be eliminated by changing the dividing direction of the mask, so the images with clearer resolutions can be displayed.

Abstract: A method for displaying three-dimensional integral images using a mask and a time division multiplexing which is configured in such a way that a three-dimensional image is displaced in a space as an element image obtained from a three-dimensional object is passed through a lenslet and a mask, the mask consisting of a blocking region through which an element image does not pass and a transmission region through which an element image passes, for thereby displaying three-dimensional images. The present invention is advantageous to play back a three-dimensional image the resolutions of which are enhanced in a depth-based integral imaging method using a time division display of an element image and a masked image.

Abstract: A method for displaying three-dimensional integral images using a mask and a time division multiplexing which is configured in such a way that a three-dimensional image is displaced in a space as an element image obtained from a three-dimensional object is passed through a lenslet and a mask, the mask consisting of a blocking region through which an element image does not pass and a transmission region through which an element image passes, for thereby displaying three-dimensional images. The present invention is advantageous to play back a three-dimensional image the resolutions of which are enhanced in a depth-based integral imaging method using a time division display of an element image and a masked image.