Browse Prior Art Database

Geometric Pattern Grading

IP.com Disclosure Number: IPCOM000075465D
Original Publication Date: 1971-Sep-01
Included in the Prior Art Database: 2005-Feb-24
Document File: 4 page(s) / 71K

Publishing Venue

IBM

Related People

Adler, RL: AUTHOR

Abstract

The present method of pattern grading can be used where various prior methods fail; namely, in the design of form fitting garments such as brassieres. It is accomplished in three steps: 1) Analytical representation of the torso 2) Universal grading 3) Approximate development of surface sections.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 40% of the total text.

Page 1 of 4

Geometric Pattern Grading

The present method of pattern grading can be used where various prior methods fail; namely, in the design of form fitting garments such as brassieres. It is accomplished in three steps: 1) Analytical representation of the torso 2) Universal grading 3) Approximate development of surface sections.

A view of the method is depicted in Fig. 1. The output of this system are cutting patterns drawn by an electronic plotter. With the proper technology, direct positioning of cutting knives to fabric may be done. The input to the system may be accomplished with a light pen and a cathode-ray tube.

The following description gives the mathematics of the above three steps from which a computer program can obviously be constructed.

Step 1. A convenient coordinate system to use, to mathematically specify human torsos is cylindrical coordinates. An origin is fixed in three dimensional space with the z-axis centrally located with respect to a particular body. A point on the surface is given by (rho, theta, z) where rho is the length of the radius from the z-axis to the point, theta is the angle between the x-axis and the radius, and z is the height of the point (see Fig. 2). The surface is specified by determining rho = rho(8, z) as a function of theta and z. This function can be experimentally determined as follows.

A dressmakers dummy is set up in an apparatus by which an angular measurement can be made as the dummy is rotated 360 degrees about a longitudinal axis. A height gauge is positioned under the axis on a flat surface where a scale is provided to measure the distance z of the gauge. The gauge itself measures the altitude of the torso. (A Sheffield Cordax measuring machine would be quite suitable for this purpose. This device provides space position readings in digital form of points on a 3-dimensional surface). Let us say that angular readings theta(i) (in radians) are taken where the index i is stepped up one unit for each degree of arc. Also position readings z(j) are taken for each 0.1 inch along the axis. The height gauge provides numerical data rho(theta(i), z(j)). This should yield more than enough data points to fit an analytical expression to the surface for a good degree of approximation. Finer measurements can be made if the situation demands. For each fixed z(j) a truncated Fourier series can be used to approximate the data with

(Image Omitted)

If symmetry assumptions are made the sine terms drop out and economies can be made in the summation for a (z(j)). A good approximation of the contour curve pi(theta, z(j)) to the data can be made by taking the value of N to be around 20. Fig. 3 is a graph of the function rho in polar coordinates.

Next, functions a(n)(z), b(n)(z) must be determined to fit the data a (z(j)) and b(n)(z(j)). For this cubic splines will provide reasonable analytic expression which can be handled in a computer program. At this point we have a function rho(theta, z) which is twice diffe...