Edge position detection device and edge position detection method

Abstract

The invention provides an edge position detection device and method. A brightness profile in an inspection image representing a group of pattern units on a substrate is acquired by an edge position detection device (71). Next, for the luminance profile (71) having four concave portions (851) to (854) arranged alternately and three convex portions, while satisfying the constraint condition based on the design data of the pattern unit group, the four concave portions ( The four bell-shaped functions corresponding to (851) to (854) and the left-right symmetrical model function (72) synthesized by the three bell-shaped functions corresponding to the three convex parts are fitted. Then, a plurality of coefficients included in each bell function of the model function (72) are determined, and edge positions of linear pattern elements are obtained based on the model function (72). As a result, edge positions can be detected with high accuracy even in inspection images acquired with relatively low resolution.

Description

technical field [0001] The present invention relates to a technique for detecting the position of an edge included in a pattern unit group in an image representing the pattern unit group. Background technique [0002] Conventionally, at the manufacturing sites of semiconductor substrates, glass substrates, printed circuit boards, etc., the width (ie, line width) of the pattern elements is measured in an image representing the linear pattern elements on the object. In such measurement, it becomes important to detect the edge position of the pattern unit with good precision. [0003] For example, in the edge position detection device of JP-A-2012-73177 (Document 1), a method of obtaining the edge position of a pattern cell with good reproducibility is proposed. In this edge position detection means, the brightness profile in the width direction of the pattern unit on the image is acquired. Next, in this luminance profile, an object position group that is a part of a pluralit...

AI Technical Summary

Problems solved by technology

[0004] However, when an image for edge position detection as described above is acquired by a camera with a relatively low resolution, the gradient of the slope portion representing the edge in the luminance profile becomes small, and it becomes difficult to detect the edge position with high accuracy.

In addition, the contrast between the background area and the pattern unit between the pattern units arranged in close proximity decreases significantly, and there is a possibility that the edge position cannot be detected.

Images