Fig 1.
The data are divided it into two indicators; namely, the geometric indicator and weight indicator. The geometric indicator was the ratio of length to the width of the egg. The second indicator was the weight indicator, which was the weight of the egg in grams.
Fig 2.
Relationship between the geometry indicator and weight indicator.
The group was classified into two clusters. The left cluster was clearly classified that there was a SY egg cluster. The right cluster was clearly classified as a DY egg cluster. The intersection area was a sample with a fuzzy member that could not be clearly identified as DY eggs or SY eggs.
Table 1.
K-means iterations.
Fig 3.
The points of SY and DY group in final iteration.
The number of SY by K-means method were 210 eggs while the number of DY were 69 eggs.
Fig 4.
A fuzzy controller scheme of DY and SY classification.
The links between four modules comprising a fuzzy controller, including a fuzzy rule base, a fuzzy inference engine, and fuzzification/defuzzification.
Fig 5.
The full triangular membership function.
Fig 6.
The half triangular membership function (Right-hand side).
Fig 7.
The half triangular membership function (Left-hand side).
Fig 8.
Input membership functions of the geometric indicator.
Fig 9.
Input membership functions of the weight indicator.
Fig 10.
Output membership functions of the truth value.
Table 2.
Fuzzy logic rules for the DY eggs detection system.
Fig 11.
Input of the geometric indicator is 1.45 mm.
Fig 12.
Input of the weight indicator is 72 grams.
Fig 13.
Result of output membership functions of the truth value.
Table 3.
Shape of the input membership function and results.
Fig 14.
A prototype of double yolk egg detection.
Fig 15.
Procedure to develop a program for a double yolk eggs detector.
Fig 16.
Interface of double yolk eggs detection.