Dimension measuring device, dimension measuring method, and semiconductor manufacturing system

A semiconductor device manufacturing system, comprising:

• a semiconductor manufacturing device in which a sample formed with an etched film is processed;
• an evaluating device in which a shape of the etched film that has undergone etching by the semiconductor manufacturing device is captured as an image;
• and
• a dimension measuring device that measures a dimension of a predetermined point in the shape of the etched film that has undergone etching by using the image captured by the evaluating device as an input image, wherein
• the processing device, the evaluating device, and the dimension measuring device are connected via a network, and
• the dimension measuring device is configured to
• generate a first image in which each region of the input image is labeled by region by machine learning,
• generate an intermediate image including a marker indicating each region of the first image based on the generated first image,
• generate a second image in which each region of the input image is labeled by region based on the input image and the generated intermediate image by an image processing based on luminance information,
• obtain coordinates of a boundary line between adjacent regions by using the generated second image,
• obtain coordinates of a feature point that defines a dimension of the predetermined point in the shape of the etched film that has undergone etching by using the obtained coordinates of the boundary line, and
• measure the dimension of the predetermined point in the shape of the etched film that has undergone etching by using the obtained coordinates of the feature point.

A dimension measuring method for measuring a dimension of a predetermined point in a shape of an etched film that has undergone etching using an image showing the shape of the etched film that has undergone etching, the method comprising:

• a step of generating a first image in which each region of the image showing the shape of the etched film that has undergone etching is labeled by region by machine learning;
• a step of generating an intermediate image including a marker indicating each region of the first image based on the generated first image;
• a step of generating a second image in which each region of the image showing the shape of the etched film that has undergone etching is labeled by region based on the image showing the shape of the etched film that has undergone etching and the generated intermediate image based on luminance information;
• a step of obtaining coordinates of a boundary line between adjacent regions by using the generated second image;
• a step of obtaining, by using the obtained coordinates of the boundary line, coordinates of a feature point that defines a dimension of the predetermined point in the shape of the etched film that has undergone etching;
• and
• a step of obtaining the dimension of the predetermined point in the shape of the etched film that has undergone etching by using the obtained coordinates of the feature point.