Research Of Detection Technology On Soft-Yolk And Cracked Preserved Egg

Preserved egg is a traditional food loved by the people and occupied an important position in our remanufactured egg industry in China, the preserved egg is mainly divided into soft-yolk preserved eggs and hard-yolk preserved eggs based on the yolk solidification, and their flavors were quite different. If it can be distinguished, that will be will be better to meet the needs of different consumers, and produce good economic benefits. Ultrasonic has good direction, penetrating, and has been widely used in the field of nondestructive testing. Ultrasonic velocity depend on elastic modulus and density of material, and ultrasonic velocity as the characteristic parameters to differentiate between the soft-yolk preserved eggs with a hard-yolk preserved eggs. During the preserved eggs process, the eggs are easy cracked due to the reasons of the production process, the operation of the workers and so on. In order to ensure the quality of the product, the damaged preserved eggs must be removed. So in this paper, a new auto method was explored to detect the crack eggs. At present, the detection techniques on the eggs are mainly based on machine vision and acoustic detection. After the eggs were cured, the surface of preserved eggs to produce a lot of stain, and this will have an image acquisition and interference, so the machine vision technology is not suitable to be used to detect preserved egg breakage. So in this paper, the preserved egg cracked detection by using acoustic detection technology. However, the preserved eggs shell was very thinning and brittle, when knocked on eggshell, it should be very careful. If the knocking power was too large, it can cause the acquisition of the sound signals are weak and the signal to noise ratiois not high. Based on the feature of the signal acquisition the wavelet analysis method was used. The main works in this paper are as follows:1. Though the test ultrasonic velocity of the soft-yolk preserved eggs and hard-yolk preserved eggs in the direction of the long-diameter and short-diameter direction, the results shown that the two parts of data overlap and it can not distinguish. Further analysis of the test results, it was found that ultrasound does not penetrate the preserved eggs, but propagates along the preserved eggs eggshell.2.The wavelet basis was determined in the preserved eggs acoustic vibration signal processing. That was quiet different from the Fourier analysis, the wavelet basis was not unique. The different results would be obtained by using different small wavelet basis. Based on the compact support, smoothness and regularity of features about the wavelet basis and the signal reconstruction ability, the best wavelet ’db6’ wavelet was selected. 3. Compared of the wavelet threshold, denoising modulus maxima denoising and translation invariant in effecting the denoising. We have found that the wavelet threshold denoising preserved eggs percussion vibration signal noise cancellation.4. the acoustic signal of nomal preserved egg and cracked preserved egg was processed with wavelet transform.then, each level wavelet decomposition of signal energy share of the total signal energy ratio of the normal preserved eggs signal energy is concentrated in the high frequency band and while damage to the preserved eggs acoustic vibration signal energy is mainly concentrated in the relatively low frequency band. The calculation of the proportion of the total energy on each decomposed signal occupied the signal, and thus constitutes the feature vector to distinguish between normal preserved eggs and crack preserved eggs.5. The6-24-2RBF neutral network was created based on MATLAB. The eigenvector based on wavelet power spectrum was formed, which was used as the input of RBF neutral network. The results was shown that this method can be used in detecting the cracks in preserved eggs.