METHOD AND DEVICE FOR ECONOMIZING COMPUTING RESOURCES TO BE USED DURING A PROCESS OF VERIFICATION OF CONVOLUTIONAL PARAMETERS USING TEST PATTERN TO ENHANCE FAULT TOLERANCE AND FLUCTUATION ROBUSTNESS IN EXTREME SITUATIONS

摘要

The present invention relates to a method and apparatus capable of saving computing resources required in the process of verifying the integrity of a convolution parameter using a test pattern. More specifically, a method of verifying the integrity of one or more parameters of a neural network and saving one or more computing resources by inserting at least one test pattern into at least one background region of at least one input image, a) When the computing device obtains information on each of the (t-2)th image and the (t-1)th image corresponding to the (t-i2)th frame and the (t-i1)th frame of the image, respectively, causing at least one background prediction unit to generate t-th background prediction information of the t-th image with reference to information on each of the (t-2)th image and the (t-1)th image; (b) the computing device, when the t-th image is obtained, causes the at least one pattern insertion unit to insert the test pattern into the t-th image with reference to the t-th background prediction information. generating an input for verification; (c) causing, by the computing device, the neural network to generate at least one output for verification by applying one or more convolutional operations to the input for verification; and (d) causing, by the computing device, at least one verification unit to determine the integrity of the neural network with reference to the verification output and at least one reference output corresponding to the test pattern; It relates to a method and apparatus comprising a.