Optics: measuring and testing – Refraction testing
Reexamination Certificate
2001-02-27
2002-05-28
Font, Frank G. (Department: 2877)
Optics: measuring and testing
Refraction testing
C356S134000
Reexamination Certificate
active
06396576
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates generally to methods for determining the location of a transitional boundary or “shadowline” between an illuminated region and a dark region on a linear scanned array of photosensitive cells, such methods being particularly applicable to critical angle refractometers wherein shadowline location is correlated to index of refraction of a test sample.
Refractometers are widely used for measuring the refractive index of a sample. In refractometers designed to measure solid and/or liquid samples, the critical angle of total reflection is measured by directing an obliquely incident convergent beam of light at a surface-to-surface boundary between a high refractive index prism and the sample and then observing a portion of the light after interaction at the boundary. In transmitted light refractometers, light that is transmitted through the sample and prism is observed, while in reflected light refractometers, the light that is reflected due to total reflection at the surface-to-surface boundary is observed. In either case, an illuminated region is produced over a portion of a detection field of view, and the location of the shadowline between the illuminated region and an adjacent dark region in the detection field of view allows the sample refractive index to be deduced geometrically. In simpler hand-held refractometers used in industry, a reticle scale is superimposed in the field of view and the user looks through an eyepiece to observe the location of the shadowline with respect to the reticle scale, which is marked so as to provide desired information such as percentage concentration of solids in the sample.
Automatic refractometers were developed to remove the guesswork associated with visually determining shadowline location with respect to a reticle scale, thus improving the accuracy (closeness to the true value) and precision (repeatability regardless of accuracy) of measurement readings. U.S. Pat. No. 4,640,616 issued Feb. 3, 1987 to Michalik discloses an automatic Abbe refractometer wherein a linear scanned array (LSA) of photosensitive elements or “cells” is arranged to detect light totally reflected at a sample/prism boundary. In a commercial embodiment, the linear scanned array includes a straight line of charge-coupled device (CCD) cells that are scanned electronically to provide a series of pulse signals each having an amplitude proportional to the amount of illumination received by the cell from incident light. Light received by the linear scanned array divides the array into an illuminated region and an adjacent dark region, thereby forming a shadowline on the array. The particular cell or interpolated inter-cell fraction at which the shadowline falls on the linear scanned array, known as the “cell crossing number,” is determined by the index of refraction of the sample substance placed in contact with the optical means. Thus, a method is required for evaluating the pulse signals from the photosensitive cells to find the cell crossing number. The cell crossing number may then be used to provide a measurement value of the index of refraction or a percent concentration of dissolved solids, such as sucrose, in the sample medium.
The Michalik patent teaches a “thresholding” approach for processing the light intensity signals from the array cells to determine the cell crossing number. This approach is represented graphically at
FIG. 3
herein. Under a thresholding approach, an empty baseline or reference scan is taken without a sample (that is, with respect to air) and stored to establish a reference illumination curve. The curve from the reference scan is then scaled by a predetermined fixed scale factor, for example 94%, to provide a threshold curve as indicated in FIG.
4
. The resulting cell where the sample scan curve intersects the threshold curve is declared the cell crossing number.
This approach yields precise and accurate measurements, however the incident light levels must be controlled to a high degree, as it is crucial that the reference and sample scans are comparable. Any deviation in intensity levels will cause erroneous readings. Since the disclosed automatic refractometer is a “reflected light” refractometer, light reaching the linear scanned array never passes through the sample, it is possible to adequately control incident light levels by controlling source luminance. An advantage of the thresholding approach is that a fresh reference scan is taken every time the instrument is turned on, so gradual changes in the response characteristics of each photosensitive cell over the lifetime of the cell (known as “response drift”) do not affect instrument performance. Moreover, use of a reference scan compensates for cell-to-cell variance in response to a given level of illumination.
The use of a reference scan and scaled threshold is problematic in “transmitted light” refractometers because, unlike the reflected light refractometer discussed above, light must pass through the sample before reaching the detector array. Consequently, sample-dependent factors such as the color, opaqueness, thickness, and homogeneity of the sample make it impractical to control incident light levels at the detector array. For example, a reference scan of air may be suitable for a clear water sample, but would not be suitable for measurement of a sample with low transmissivity, such as ketchup. Another drawback of the threshold approach is that a single defective cell providing an erroneous response signal may, at the worst, distort the measurement result.
U.S. Pat. No. 5,617,201 issued Apr. 1, 1997 to Kåhre describes a reflected light refractometer using another method for determining the cell crossing number of the shadowline boundary. The method involves describing the illumination distribution curve by means of a mathematical model, and using the mathematical model to find the cell crossing number. In the described embodiment, the illumination distribution curve is represented by three different straight lines A, B, and C representing a light region of the array, a transition region of the array from light to dark, and a dark region of the array, respectively. The intersection of line B with line C is chosen as the cell crossing number. Non-linear models are also suggested. A similar approach with respect to a transmitted light refractometer is taught in U.S. Pat. No. 6,172,746 issued Jan. 9, 2001 to Byrne et al. (this patent shares a common assignee with the present application), and is illustrated herein at FIG.
5
. However, the method involving intersecting “best fit” straight lines representing the transitional and dark regions of the detector array proved to be too imprecise for transmitted light applications wherein detected light levels are difficult to control. The precision attained was inadequate over the range of light intensities that this instrument can experience, and thus the method was ultimately not adopted.
BRIEF SUMMARY OF THE INVENTION
Therefore, it is a primary object of the present invention to provide a new method for determining the cell crossing number of a shadowline on a photosensitive array that is relatively immune to light intensity level variations while maintaining adequate precision.
This object is achieved by a method that is now briefly described in a preferred form. Initially, the photosensitive array is scanned to extract a response signal from each of the photosensitive cells in the array that represents the amount of illumination of the corresponding cell by incident light. The response signals from the photosensitive cells are converted from analog form to digital pixels, thus yielding a set of data points that collectively represent an illumination distribution curve over the array. A range of cells within which the shadowline resides is established by analyzing the illumination curve data. A preferred procedure for establishing a “start” cell for this range is to find the brightest cell by looking for a peak pixel value, and then step forward one cell at a time until a cell
Font Frank G.
Leica Microsystems Inc.
Nguyen Tu T.
Simpson, Simpson & Snyder, PLLC
LandOfFree
Method for determining shadowline location on a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for determining shadowline location on a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for determining shadowline location on a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2899578