Produce color data correction method and an apparatus therefor
View Patent ↗A produce recognition system comprises an image capture device arranged to (i) capture a first color image which is representative of a color image of a produce item, and (ii) capture a second color image which is representative of a color image of at least one target color swatch. The produce recognition system further comprises control circuitry arranged to (i) calculate one or more color correction factors based upon differences between the captured second color image and a store of reference color images, and (ii) apply the calculated one or more color correction factors to the captured first color image to correct for color variations in the color image of the produce item due to a combination of variations in natural lighting and variations in interior lighting.
1. A produce recognition system comprising:
an image capture device arranged to (i) capture a first color image which is representative of a color image of a produce item, and (ii) capture a second color image which is representative of a color image of at least one target color swatch; and
control circuitry arranged to (i) calculate one or more color correction factors based upon differences between the captured second color image and a store of reference color images, and (ii) apply the calculated one or more color correction factors to the captured first color image to correct for color variations in the color image of the produce item due to a combination of variations in natural lighting and variations in interior lighting.
2. A produce recognition system according to claim 1 , wherein the control circuitry arranged to calculate one or more color correction factors includes the control circuitry arranged to generate a transformation matrix based upon a comparison of colors between the captured second color image and the store of reference color images.
3. A produce recognition system according to claim 2 , wherein the control circuitry arranged to calculate one or more color correction factors includes the control circuitry arranged to apply a back propagation algorithm during generation of the transformation matrix.
4. A produce recognition system according to claim 3 , wherein the control circuitry arranged to apply the calculated one or more color correction factors includes the control circuitry arranged to apply a back propagation algorithm during application of the calculated one or more color correction factors to the captured first color image to correct for color variations in the color image of the produce item due to the combination of variations in natural lighting and variations in interior lighting.
5. A produce recognition system according to claim 1 , further comprising a weighing scale having a weighing plate in which the at least one target color swatch is embedded.
6. A produce recognition system according to claim 5 , wherein the image capture device has an aperture with a field of view in which the at least one target color swatch lies.
7. A produce recognition system according to claim 5 , wherein the at least one target color swatch comprises four target color swatches embedded in the weighing plate of the weighing scale.
8. A produce recognition system according to claim 7 , wherein the image capture device has an aperture with a field of view in which each of the four target color swatches lies.
9. An apparatus for correcting variations in measured produce color in a produce recognition system, the apparatus comprising:
at least one target color swatch;
a store of reference color swatches;
an image capture device arranged to (i) capture first color data which is representative of a color image of a produce item, and (ii) capture second color data which is representative of a color image of the at least one target color swatch illuminated by a combination of natural light and interior light; and
control circuitry arranged to (i) compare the captured second color data with a store of reference color data which is representative of the store of reference color swatches, (ii) calculate one or more color correction factors based upon the comparison of the captured second color data with the store of reference color data, and (iii) process the first color data based upon the calculated one or more color correction factors to provide corrected first color data which is representative of a corrected color image of the produce item.
10. An apparatus according to claim 9 , wherein the at least one target color swatch comprises an array of 24 scientifically prepared natural, chromatic, primary, and grayscale color squares in a range of different colors.
11. An apparatus according to claim 9 , wherein the store of reference color swatches comprises a store of color image data which is representative of at least one reference color swatch against which color image data which is representative of the at least one target color swatch is compared.
12. An apparatus according to claim 9 , wherein the image capture device comprises a charge coupled device (CCD).
13. A method of correcting variations in measured produce color in a produce recognition system, the method comprising:
capturing first color data which is representative of a color image of a produce item;
capturing second color data which is representative of a color image of one or more target color swatches illuminated by a combination of natural light and interior light;
comparing the captured second color data with a store of reference color data;
calculating one or more color correction factors based upon the comparison of the captured second color data with the store of reference color data; and
processing the first color data based upon the calculated one or more color correction factors to provide corrected first color data which is representative of a corrected color image of the produce item.
14. A method according to claim 13 , wherein the first and second color data are captured at the same time.
15. A method according to claim 13 , wherein capturing second color data includes capturing color data which is representative of a color checker.
16. A method according to claim 15 , wherein comparing the second color data with a store of reference color data includes comparing the captured color data which is representative of a color checker with the store of reference color data.
17. A method according to claim 16 , wherein calculating one or more color correction factors includes generating a transformation matrix based upon the comparison of the captured color data which is representative of a color checker with the store of reference color data.
18. A method according to claim 17 , wherein processing the first color data includes applying a back propagation algorithm to the generated transformation matrix to provide the corrected first color data which is representative of the corrected color image of the produce item.