Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

2results about How to "Simple layer" patented technology

Methods for compensating for overprinting deviation

A method for compensating overlay misalignment includes: providing a wafer; performing a first pre-compensation and a second pre-compensation on a lithography machine; forming an initial fin structure on the wafer, wherein the sum of the compensation value of the first pre-compensation and the previous first etching misalignment of the initial fin structure is 0; obtaining a first etching misalignment of the initial fin structure, wherein the first etching misalignment is the same as the compensation value of the second pre-compensation; performing a third compensation on the lithography machine based on the first etching misalignment; performing a fourth pre-compensation on the lithography machine; after performing the third and fourth pre-compensations on the lithography machine, partially etching the initial fin structure to form a fin structure, wherein the sum of the compensation value of the fourth pre-compensation and the compensation value of the second pre-compensation is 0, and the compensation value of the second pre-compensation is the same as the previous second etching misalignment of the fin structure; obtaining a second etching misalignment of the fin structure, wherein the value of the second etching misalignment is 0. This method improves overlay accuracy.
Owner:SEMICON MFG INT (SHANGHAI) CORP

Millimeter-wave contraband detection method based on multi-head attention and BIFPN to improve YOLOv7

The present invention claims a millimeter wave contraband detection method based on multi-head attention and BIFPN improved YOLO v7 algorithm, belonging to the field of image processing technology. The steps of the method are as follows: S1: Use millimeter wave radar to actively capture target echoes, then use the range migration algorithm (RMA) to reconstruct the millimeter wave three-dimensional image of the target, and obtain a two-dimensional millimeter wave image by projecting upward at a distance; S2: Input the millimeter wave image into the trained improved YOLO v7 network for image preprocessing and feature extraction; the improved YOLO v7 network introduces multi-head attention and BIFPN weighted bidirectional feature pyramid network; S3: Perform contraband target detection and recognition based on feature extraction information such as target position and category obtained by the network output layer; S4: Automatically generate annotated contraband detection results, including target category, coordinates and confidence, based on the obtained contraband detection and recognition information. Compared with traditional contraband detection methods and traditional millimeter wave image recognition methods, this system has the advantages of fast training speed, high efficiency, high result resolution, and high recognition accuracy.
Owner:CHONGQING UNIV OF POSTS & TELECOMM