Abstract: A lithography apparatus includes a formation unit that forms an alignment mark on a substrate by irradiating the substrate that includes a photosensitizer with light, and a transfer unit that aligns the substrate on the basis of the position of the alignment mark and that transfers a pattern to the substrate by illuminating the photosensitizer with exposure light. The formation unit irradiates a material of a grounding of the photosensitizer with irradiation light at a wavelength that differs from that of the exposure light and forms the alignment mark on the material by processing the material with energy of the irradiation light.

Abstract: A lithography apparatus includes a formation unit that forms an alignment mark on a substrate by irradiating the substrate that includes a photosensitizer with light, and a transfer unit that aligns the substrate on the basis of the position of the alignment mark and that transfers a pattern to the substrate by illuminating the photosensitizer with exposure light. The formation unit irradiates a material of a grounding of the photosensitizer with irradiation light at a wavelength that differs from that of the exposure light and forms the alignment mark on the material by processing the material with energy of the irradiation light.

Abstract: A position detection method of detecting a position of a detection mark. The method includes a matching step of calculating a value of a correlation using a template for an image including the detection mark, a change step of changing the template to be used for the correlation, a repeat step of, when a value of a correlation in the matching step is not more than a predetermined value, changing the matching processing while changing the template in the change step, and a step of performing position detection on the basis of a result of correlation in the matching step or the repeat step.

Abstract: A position detection method of detecting a position of a detection mark. The method includes a matching step of calculating a value of a correlation using a template for an image including the detection mark, a change step of changing the template to be used for the correlation, a repeat step of, when a value of a correlation in the matching step is not more than a predetermined value, changing the matching processing while changing the template in the change step, and a step of performing position detection on the basis of a result of correlation in the matching step or the repeat step.

Abstract: A method of detecting a position of a mark within image data. The method includes a first step of obtaining a first position of the mark based on a first degree of correlation between a first template and the image data, the first template having a first feature point to obtain the first degree of correlation, a second step of obtaining a second position of the mark based on a second degree of correlation between a second template and the image data, the second template having a second feature point, of which a position is different from that of the first feature point, to obtain the second degree of correlation, and a third step of detecting the position of the mark through the first and second steps.

Abstract: Especially, a prealignment mark on a wafer looks in various ways as the density of semiconductor devises increases. As reasons, the line width changes due to the multi-layered structure of the mark portion, or the brightness changes due to the difference in reflectance of the mark portion. To improve the alignment mark detection ratio even in such an image, the edges of the mark and their directions are simultaneously extracted. Matching with a template stored in advance and formed from feature points of interest, which represent the edge position of each edge direction, is performed. At this time, the template is automatically deformed in accordance with deformation of the mark and a change in brightness of the mark, thereby optimizing the template. Self-learning is executed such that the optimized template is to be used for matching of next and subsequent times. With this processing, the mark detection ratio increases.