초록 ▼

Fabric wrinkles are automatically evaluated using a reliable, accurate, affordable, and efficient system. Two algorithms are used, the facet model algorithm and the plane-cutting algorithm, to extract features for evaluating wrinkles in fabrics. These algorithms eliminate the need for independent ev

Fabric wrinkles are automatically evaluated using a reliable, accurate, affordable, and efficient system. Two algorithms are used, the facet model algorithm and the plane-cutting algorithm, to extract features for evaluating wrinkles in fabrics. These algorithms eliminate the need for independent evaluation of a fabric specimen by a technician.

대표청구항 ▼

What is claimed is: 1. An active imaging system for determining the smoothness of the surface of a fabric, the system comprising: a support for holding the fabric in a plane; a single-line laser projector positioned at an angle above said plane, the single-line laser projector capable of directing

What is claimed is: 1. An active imaging system for determining the smoothness of the surface of a fabric, the system comprising: a support for holding the fabric in a plane; a single-line laser projector positioned at an angle above said plane, the single-line laser projector capable of directing radiation onto at least one portion of the surface of the fabric; a single range camera positioned above said plane, the range camera capable of capturing at least one range image of the at least one portion of the surface of the fabric based on the radiation from only the single-line laser projector, the range image including a topographic map including height differences of a plurality of points on the surface of the at least one portion of the surface of the fabric; and a processor that obtains the range image, and determines the smoothness of the surface of the fabric using an algorithm that incorporates one of a facet model and a plane-cutting method, wherein, the smoothness of the surface is determined by: generating a plurality of parallel planes that intersect the surface at a plurality of range levels, recording an intersection area with the surface for each of the parallel planes, calculating a category of a plurality of categories for each of the intersection areas, the plurality of categories including an area under the curve of the intersection area versus the range level, and determining a degree of smoothness for the surface. 2. The system of claim 1, wherein the range camera is a CMOS camera. 3. The system of claim 1, wherein the support is a platform with a planar surface for supporting the fabric. 4. The system of claim 3, wherein the planar surface is parallel to the optics plane of the range camera. 5. The system of claim 3, wherein the planar surface is perpendicular to a line of sight of the range camera. 6. The system of claim 1, wherein the range camera is at least 21 inches above the support. 7. The system of claim 1, wherein the single-line laser projector is capable of directing radiation at the support along a line of sight approximately 80 degrees away from a line of sight of the range camera. 8. The system of claim 1, wherein the single-line laser projector, the sensor and the support are arranged substantially in a triangle. 9. The system of claim 1, wherein the range camera moves over the support to capture the range image of the surface of the fabric. 10. The system of claim 1, wherein the support moves along a plane parallel to an optics plane of the range camera. 11. The system of claim 1, wherein the range camera captures the at least one range image based on radiation, from the projector, that is reflected from the surface of the fabric. 12. The system of claim 11, wherein the topographic data comprises at least one height of a point in the surface of the fabric relative to the at least one reference point. 13. The system of claim 1, wherein the processor preprocesses the range image by enhancing wrinkles on the surface of the fabric. 14. The system of claim 1, wherein the smoothness of the surface is determined by approximating the surface using a facet model to provide a piece-wise continuous surface of connected facets; calculating a category of a plurality of categories for each of the connected facets, the plurality of categories including at least one of slope, ridge, peak, ravine, and pit; and determining a degree of smoothness for connected facets that are proximate to each other. 15. The system of claim 1, wherein the processor extracts features on the surface to quantify the smoothness. 16. A method for detecting wrinkles in a surface of a fabric using a topographic map of the surface to determine a smoothness of the surface of the fabric, the method comprising: positioning a single-line laser projector at an angle over a plane supporting the fabric; directing radiation onto at least one portion of the surface of the fabric from only the single-line laser projector; obtaining a range image of the at least one portion of the surface of the fabric based on the directed radiation by using a single range camera, the range image including a topographic map including height differences of a plurality of points on the surface of the at least one portion of the surface of the fabric; and determining the smoothness of the surface of the fabric using an algorithm that incorporates one of a facet model and a plane-cutting method, wherein, the smoothness of the surface is determined by: generating a plurality of parallel planes that intersect the surface at a plurality of range levels, recording an intersection area with the surface for each of the parallel planes, calculating a category of a plurality of categories for each of the intersection areas, the plurality of categories including an area under the curve of the intersection area versus the range level, and determining a degree of smoothness for the surface. 17. The method of claim 16, wherein the processor preprocesses the range image by enhancing wrinkles on the surface of the fabric. 18. The method of claim 16, wherein the smoothness of the surface is determined by approximating the surface using a facet model to provide a piece-wise continuous surface of connected facets; calculating a category of a plurality of categories for each of the connected facets, the plurality of categories including at least one of slope, ridge, peak, ravine, and pit; and determining a degree of smoothness for connected facets that are proximate to each other. 19. The method of claim 16, wherein the processor extracts features on the surface to quantify the smoothness. 20. A method for detecting wrinkles in a surface of a fabric using a topographic map of the surface to determine a smoothness of the surface of the fabric, the method comprising: obtaining a range image of at least one portion of the surface of the fabric, the range image including a topographic map including height differences of a plurality of points on the surface of the at least one portion of the surface of the fabric; and determining the smoothness of the surface of the fabric using an algorithm that incorporates one of a facet model and a plane-cutting method, wherein, the smoothness of the surface is determined by: generating a plurality of parallel planes that intersect the surface at a plurality of range levels, recording an intersection area with the surface for each of the parallel planes, calculating a category of a plurality of categories for each of the intersection areas, the plurality of categories including an area under the curve of the intersection area versus the range level, and determining a degree of smoothness for the surface.