Chip image super-resolution reconstruction method based on depth learning

Abstract

The invention discloses a chip image super-resolution reconstruction method based on depth learning, which mainly solves the problem of low resolution at the dense part of the circuit of the chip image reconstruction in the prior method. The technical scheme is as follows: 1. dividing the image set and constructing the training data set; 2. training the training data set; 3. estimating the sub-pixel displacement of K low-resolution images and reference images; 4. carrying out up-sampling on that reference image, inputting the reference image into the train model, and outputting the estimated image; 5. degrading the estimated image, and calculating the simulation error between the degrade image and the K low-resolution images; 6. superimposing the simulation error on the estimated image toobtain an improved estimate image; 7. Iteration being performed on steps 5 to 6 until that error function is less than the error threshold, and the resulting improved estimated image being output. Theinvention improves the super-resolution reconstruction effect of the circuit dense part in the chip image, and can be used for detecting the hardware Trojan horse at the circuit dense part of the chip.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to an image super-resolution reconstruction method, which can be used for hardware Trojan horse detection at chip-intensive circuits. Background technique [0002] At present, image super-resolution reconstruction technology plays an important role in improving the resolution of chip images. In recent years, my country's semiconductor industry has developed rapidly, but some key parts of high-end chips still rely on imports, and my country's integrated circuit design and manufacturing technology is not perfect, so the hardware Trojan horse problem introduced in the chip design and production process cannot be ignored , the hardware Trojan horse refers to a tiny malicious circuit lurking in the original circuit. Under special conditions, the module can change the function of the circuit, leading to serious consequences of information leakage or even destroying the syste...

AI Technical Summary

Problems solved by technology

However, the microscope and camera equipment required to take high-resolution chip photos are very expensive. In order to reduce costs, people usually use ordinary cameras to take images, and there are external interference factors such as atmospheric disturbances, light changes, and noise. These image degradation factors make The image resolution we acquire is usually low, so we need to use image super-resolution reconstruction technology to improve the resolution of the chip image

[0003] In terms of research methods, image super-resolution reconstruction techniques can be divided into three categories: interpolation-based, reconstruction-based, and learning-based. Interpolation-based methods generally have obvious jagged effects, and reconstruction-based methods take into account the degradation model of the image, and can Combining the prior knowledge of the image, the performance has been greatly improved compared with the interpolation method, but the effect is still not good when applied to the chip image; the main idea of ​​the learning-based super-resolution algorithm is to learn the difference between the low-resolution image and the high-resolution image. According to the corresponding relationship between them, the super-resolution reconstruction of the image is guided. With the rise of machine learning, the super-resolution reconstruction algorithm based on deep learning gradually emerges. When dealing with ordinary natural images, these methods show It has excellent performance, but when reconstructing the chip image composed of dense circuits, it cannot handle the details of the image very well.

Images