Color Classification of Non-Uniform Baked and Roasted FoodsRobert K. McConnell, Jr., Henry H. Blau, Jr.
Based on paper presented at:
FPAC IV Conference
3-5 November 1995
AbstractAlthough the color of baked and roasted materials is used as an indication of quality or degree of cooking, the nature of many of these materials renders them ill-suited for traditional color based automated classification. Here we identify some of the problems and show how off the shelf software, using minimum description based color classification can be used to produce results generally in good agreement with "subjective" human classification.
IntroductionAs long as food has been baked or roasted, color has presumably been an important guide to whether it has been properly cooked. Even if cooking is adequate, if the color does not conform with the customer's preconceptions the chances of a purchase are greatly diminished.
The advantages of having reliable, automated, full color inspection systems whose decisions are consistent with human perception have long been recognized. Now that inexpensive color cameras and color frame grabber boards are readily available, the major obstacle to widespread use of color has been the lack of suitable general classification methods.
To mimic human color recognition a machine vision system must be able to operate in a single three-dimensional color space, learn quickly, and handle multi-modal, overlapping, color distributions easily. At the same time it must remain sensitive to anomalous characteristics, even under non-uniform lighting conditions.
Typical Applications in Baking and Roasting
Minimum Description AnalysisThe theoretical basis for use of minimum description analysis for color based classification is discussed extensively elsewhere (see for example McConnell and Blau, 1992, 1994, McConnell, Massa and Blau, 1995). This simple, general approach provides particularly efficient color classification using only a reference histogram, based on the color distribution of each class of interest and a test histogram, representing the color distribution of the object to be classified. We measure the dissimilarity of distributions for a variety of likely reference color distributions and then choose the reference class with the smallest measured dissimilarity.
The approach can be shown to be closely related to a Bayesian maximum likelihood classification and usually produces results in good agreement with those of human inspectors.
The system is trained much like a human by simply showing it examples of the various classes of interest from which it builds the reference histograms. Training time is typically of the order of a few seconds. The system can then be placed in classification mode where it performs the inspection and reports the results with little or no human intervention.
This classification method is now being successfully used in a wide variety of inspection and process control applications the electronic, automotive, food products and recycling industries. The results generally coincide well with those made by a human inspector.
Muffin Doneness ClassificationWe trained WAY-2C, our minimum description based color classification system, on each of the reference muffins in Fig. 1a using a square region of interest approximately equal in area to the muffin top. The training region included examples of cake, blueberries, shadow and foil. When shown the test muffin in Fig. 1b the system classified it as being most similar to reference muffin #4.
Peanut Color MatchingTwo separate sets of tests were performed on the peanuts.
For the first tests we poured approximately 200g of each sample into a 10x10cm rectangular plastic box. We then trained WAY-2C using four different images of each of samples #65 and #53. Between acquisition of each image the sample was rotated. At least once during the process it was mixed and then gently shaken to level it. Training time for the complete set was under 3 minutes.
When the training was complete we repeatedly classified each of the three samples, rotating, shaking, and stirring them from time to time to determine the effect of these actions.
The second set of tests used moving samples. For this set we used a rotating turntable containing approximately 500g of peanuts in a 22cm diameter pie plate. The training and inspection area was the same size as that used in the stationary tests.
Because of the color variations at scales of several inches, even though each of our measurements sampled an area of about 60 square centimeters, fluctuations were still large enough for occasional regions in sample #53 to most resemble sample #65 and vice versa.
In spite of the occasional fluctuations, the results of all the tests consistently indicated that the surface color distribution of the unknown sample more closely resembled that of sample #53 than sample #65.
After hearing that laboratory experiments by the sample provider based on colorimetry found the "unknown" sample more closely resembled Sample #65 we repeated the original tests, did some careful quantitative interpolation, and even had seven different unbiased human observers visually rank the samples in order of color under a variety of artificial and natural lighting conditions. All of these agreed with our original results: the "unknown" sample most resembles Sample #53. In other words the minimum description approach was in agreement with the human classification and in disagreement with a method which utilized the mean color of the peanuts.
ConclusionsBased on the above results, and the performance in similar applications where the system is in routine operation, one can conclude that minimum description color classification is well suited to classifying of baked and roasted foods similar to those tested here.
References1. McConnell, R.K. and H.H. Blau. 1992. A powerful, inexpensive approach to real-time color classification. Proceedings Soc. Mfg. Engs. Applied Machine Vision Conference '92, June 1-4, 1992, Atlanta, SME Technical Paper MS92- 164, Society of Manufacturing Engineers, Dearborn, Michigan.
2. McConnell, R.K. and H.H. Blau. 1994. Minimum description classification: a new tool for machine vision color inspection, Proc. FPAC III Conference, February 9-12, 1994, Orlando, Amer. Soc. Agr. Engs., St. Joseph, Michigan.
3. McConnell, R.K., R.A. Massa and H.H. Blau. 1995. Color machine vision. Proceedings Sensors Expo Boston, May 16-18, 1995.