331 F. 2d 894 - Application of Traver J Smith

This appeal is from the decision of the Patent Office Board of Appeals affirming the rejection of claims 8, 9, and 19-28 of application serial No. 494,783, filed March 16, 1955, entitled "Color Grading." No claims are allowed.

The invention relates to color grading of substances such as food products by measurement of light reflection and particularly to alleged improvements in the nature of the light source and the method for indicating the degree of light reflection. Broadly, such color grading is acknowledged by appellant to be old and is based on well-known principles.

Claim 28 is directed to apparatus, all other claims to method. Appellant's apparatus, in which his method is carried out, is housed in a cabinet with a hole in the top into which a transparent glass beaker containing the sample to be tested can be set. When in place, the beaker rests on a transparent plate underneath which is a tubular light shield spaced somewhat below the plate. Surrounding the shield is a tubular light source, such as a neon or mercury vapor lamp. In a chamber beneath the shield is a photoelectric cell, and a glass light filter can be positioned to determine what portion of the reflected light coming from the sample to be tested will reach the cell. The light from the source thus illuminates the sample which reflects light down through the shield and the filter and the light reaching the photoelectric cell causes a current which is proportional to the amount of light striking the photocell cathode. This relatively weak current is amplified by an amplifier and the meter which is included in the amplifier circuit shows its value.

Appellant claims his invention to reside in two improved features: (1) providing "substantially monochromatic light" and (2) calibrating the ammeter to provide the maximum range for readings between two extremes of sample color intensity.

As to feature (1), appellant discloses as one light source a mercury vapor tube which, when energized, emits light of discrete wave lengths, viz., 404.1, 435.8 and 546.1 millimicrons. All but one wave length he says is removed by his filter so that light of a single wave length strikes the photocell. Other gas discharge tubes may be used, e. g., neon tubes, the advantage of all such tubes being that they emit relatively small numbers of discrete wave lengths as compared with incandescent light bulbs which emit a broad spectrum of radiation.

As to appellant's calibration method (2), it consists of two steps: setting the meter to read zero when the darkest-color sample is tested by adjusting a first potentiometer suitably placed in the circuit; and setting the meter to read at or near maximum when the lightest-color sample is tested by adjusting a second potentiometer. Thus, regardless of the degree of difference in color between samples at two extremes, readings indicating that difference can be spread out over the full scale of the meter.

"8. The method of color grading a product by measuring the lightness of the product by employing a source of substantially monochromatic light, said monochromatic light being adapted to be reflected from the product that is to be graded and the intensity of said deflected monochromatic light being characteristic of a condition of the product being graded comprising the steps of determining color standards relative to the opposite ends of the color range of the product to be graded, providing an electrical system including, an amplifier, a light sensitive cell and a meter, calibrating the meter with reference to the darker of the two samples to determine a low reading limit on the meter, employing the lighter colored sample to fix an upper reading of the meter, and thereafter exposing a sample of the product to produce reflected light therefrom to produce an electric signal from this reflected light to provide a reading on the meter as a direct indication of the color grade of the article.

"28. Apparatus for grading a product to determine a condition such as the completeness of processing or ripeness of the product in which such condition may be determined from the intensity of monochromatic light reflected from the product, said substantially monochromatic light being selected so that the intensity of its reflection from the product provides a measure of the condition of the product, the improvement comprising means for producing substantially monochromatic light, said substantially monochromatic light being in the portion of the light spectrum in which the condition of the product is to be determined, means for directing said substantially monochromatic light first to illuminate a sample of the product that is in the desired color condition and second to illuminate a sample of the product that is in the least desired acceptable color condition indicating means having a scale for indicating the intensity of the reflected substantially monochromatic light from said product samples, said indicating means including means to provide a first indication on said scale corresponding to the intensity of the light reflected from the desired product and a second indication corresponding to the intensity of the light reflected from the least desired acceptable product, means for adjusting said indicating means so that said first indication appears substantially on one end of said scale and for adjusting said indicating means so that said second indication appears substantially on the other end of said scale, thereby to provide a range between the ends of said scale between which various degrees of the color condition of the product being determined may be indicated so that random samples of the product may be illuminated by said substantially monochromatic light and a direct reading of the color condition thereof obtained from indications produced on said scale."

Claims 9 and 19-27, all method claims, differ essentially from claim 8 as follows: claim 9 states that only a "portion" of the sample be tested; claim 19, that light "include" monochromatic light; claims 20, 21 and 22, that the light be of a specific wave length. Claims 23-27 for purposes of this opinion are substantially the same as claim 19.

The sole issue is obviousness of the claimed improvements in view of the following prior art:

  Lehde       2,287,808 June 30, 1942
Lukens et al. 2,701,502 Feb. 8, 1955

Powers et al., "Electronic Color Sorting of Fruits and Vegetables", Agricultural Engineering, Vol. 34, No. 3; March 1953, pages 149-154, 158.

The examiner rejected all method claims as "unpatentable over either Lukens et al or Powers et al, either reference taken alone or either in view of the other," and as "unpatentable in view of the disclosure in Lehde * * *." Apparatus claim 28 was rejected "over Lukens et al or Lehde each taken separately." The board sustained the examiner on all grounds of rejection. We find it unnecessary to discuss the Lehde reference.

Lukens et al. describe a photometric food-grading device wherein a light source is used in conjunction with red and blue light filters to grade corn on the cob, judging its maturity by color. The difference in its reflectance of red and blue light is measured by a photocell and amplifier circuit similar to appellant's, such difference in reflectance being taken as a measure of corn color. Lukens et al. set the zero meter reading with a first potentiometer, and use a second potentiometer "for adjustment of the range of readings to be obtained for a given range of [corn] maturity." Lukens et al. describe their light source simply as a "light bulb," appellant interpreting this to be a conventional incandescent light. With respect to using one color (monochromatic) light rather than red and blue light alternately, Lukens et al. state:

"Simplified devices have been proposed in which the object is irradiated with light of a distinct color and a measurement made, through a photoelectric tube and electronic componets, of the amount of light reflected. Devices of this type are suitable in some applications but oftentimes the spread in reflectance between an immature and a mature product is simply not sufficient to yield reliable results."

The Powers et al. article relates to an experimental color sorter for lemons by which lemons are sorted into five commercial color classes. The reference describes, inter alia, three methods for making reflectance measurements, "Method I" being most pertinent. We quote the essence of "Method I":

"The simplest way to classify a lemon as to color is to characterize its reflectance curve by a single measurement of reflectance. It is evident that this measurement should be made in the region of the spectrum where the change in reflectance between consecutive color classes is greatest. The curves in Fig. 3 dip sharply in the vicinity of 6,780 A, where chlorophyl has an absorption peak. The variation of reflectance with lemon maturity is, clearly, a maximum at this wave length. Hence the reflectance of the fruit at 6,780 A is chosen as a criterion of color.

"To measure this reflectance, the fruit is illuminated with light restricted to a narrow band of wave lengths in the vicinity of 6,780 A. A portion of the reflected light is intercepted by a photocell. The resulting photoelectric current is proportional to the reflectance of the lemon." [Our emphasis.]

The examiner's position is that the steps recited in the method claims "are substantially identical with the steps disclosed by Lukens et al, and that the test conditions set forth in the first few lines of the claim are also anticipated in the Lukens et al patent." With respect to the use of "monochromatic light" the examiner said, "applicant's light source is not actually monochromatic and his device requires a color filter 18 to get the proper results. * * * Also, Lukens et al teaches * * * that a single color of light has been used in the art, and it is thought it would be obvious to use a single `color' in Lukens et al."

As to Powers et al. the examiner said that "Method I" considered with the rest of the reference "appears to anticipate" the method claims. As for calibrating the ammeter, the examiner said "it appears that [the calibrating] steps * * merely involve obvious steps in calibrating an indicating instrument and it seems that a person versed in the art of using instruments of the type described by Powers et al would arrange the photoelectric circuit so that the lowest readings i. e. for green lemons would come near one end of the scale and the `high' readings near the other end of the scale."

Apparatus claim 28 was rejected by the examiner thus: "Since apparatus claim 28 defines essentially the same subject matter as claim 8, merely substituting `means' for the `steps' in claim 8, said claim 28 is thought to be unpatentable for the same reasons advanced against claim 8, supra."

The board affirmed all the examiner's grounds of rejection, relying specifically on Lukens et al. alone but saying, "The further rejections based on the Powers et al. and Lehde patents appear to be merely cumulative and will be sustained without additional discussion."

Appellant argues that the board erred in holding that Lukens et al. teach the use of monochromatic light since the patentees state the use thereof does not "yield reliable results"; that Lukens et al. do not teach how to "modify their apparatus to use a distinct color"; that as to calibration of the instrument, the board "committed error because of a misunderstanding of the facts"; that Lukens et al. teach calibration of the lower end of a meter but not the upper end; and that three affidavits of record purporting to show commercial success of the invention should weigh heavily in appellant's favor.

Notwithstanding appellant's arguments, we find no reversible error in the board's decision. As for the use of "monochromatic light," "Method I" of Powers et al., quoted in part supra, is a clear teaching that light of a single wave length (6,780 A) or restricted to a narrow band of wave lengths (in the vicinity of 6,780 A) may be used in reflectance testing.

With respect to calibration of the ammeter, we agree with the examiner that "a person versed in the art of using instruments of the type described by Powers et al would arrange the photoelectric circuit so that the lowest readings * * would come near one end of the scale and the `high' readings near the other end * * *." In any event, we think Lukens et al. recognized such desideratum in stating, "Potentiometer 56 is provided to allow for adjustment of the range of readings to be obtained for a given range of maturities." (Our emphasis.) When appellant sets his high reading near or at the top of his meter scale he is merely adjusting for the greatest range possible. We think doing this would be obvious to one of ordinary skill in this art.

We have considered the affidavits of record, but agree with the solicitor that the "showing * * * [is] not * * * persuasive of patentability in the absence of reasonable doubt concerning the merit of the examiner's rejections."

The decision of the board is affirmed.

Affirmed.