• Journal of Biosystems Engineering
• /
• 제23권6호
• /
• pp.621-628
• /
• 1998

In order to develop an inspection algorithm for an automatic eggshell inspection system, acoustic impulse response with neural network method was studied. An improved error backpropagation algorithm was selected as a loaming rule of neural network, and three layer network was chosen for the neural network architecture. Acoustic signals in time domain and theirs power spectrum were studied as the input to the neural network. The classification feasibility and success rate were investigated in terms of statistical analysis and neural network approach. As a result, the success rate was 95% with the statistical model having five independent variables. Among the neural network models studied, the power spectrum of acoustic signal as the input with 64 input neurons and the two impact data showed the success rate of 95.5% which was slightly higher than of statistical analysis.

• Journal of Biosystems Engineering
• /
• 제23권1호
• /
• pp.67-74
• /
• 1998

A nondestructive quality inspection technique using acoustic impulse response method was developed for eggshell inspection. An experimental system was built to generate the impact force, to measure the response signal and to analyze the frequency spectrum. This system includes an impulse generating unit, an egg holding seal a microphone with preamplifier, and a DSP board installed on Personal Computer. A simple algorithm .was developed for crack detection. Using the developed system with algorithm, crack detection ability was evaluated and the error rate to estimate the normal egg as cracked was found to be 4% and the error rate to estimate the cracked egg as normal was also found to be 4%. This system could be adopted in industry with some modification.

• Journal of Biosystems Engineering
• /
• 제24권3호
• /
• pp.253-258
• /
• 1999

This study was conducted to develop an automatic egg inspection system for detecting creaked eggs based on acoustic impulse response. This system includes a sound generator, a sound sensor with signal conditioner, and a computer. The sound generator that hit the sharp of the dull edges of an egg was constructed with a ceramic ball pendulum attached to a rotary type solenoid. The signal conditioner included a pre-amplifier and a digital signal processing (DSP) board. The parameters for distinguishing cracked and normal eggs were the area, the geometric centroid and the resonance frequency of power spectrum of the acoustic signal generated. An algorithm for on-line detection of the continuous transferring eggs was developed. The performance tests resulted with 91% success rate to separate cracked and normal eggs at the rate of 1 second per an egg.

• 한국식품위생안전성학회지
• /
• 제17권3호
• /
• pp.129-136
• /
• 2002

본 조사는 계란 유통실태 및 위생상태를 파악하여 품질향상과 위생관리에 도움이 되는 자료를 제공하고자, 2001년 3~4개월에 인천광역시 5개 백화점 식품매장에서 총 57종의 계란을 구입하여 조사하였다. 계란의 포장은 10개 이하의 소량 단위포장 비율이 높았고, 전 품목을 실온에서 보관하였다. 유통기한은 40종(70.2%)이 28일 이상으로 표기하였다. 유통되는 특수란은 31종(54.4%) 이었다. 유통기한과 산란일자는 모든 품목에서 표시되었으며 주로 상표나 난좌에 표기되었다. 내용량에 대한 표시는 38종(66.6%)만이 실시하였다. 난각에 분변이나 털이 있는 경우 27건(9.5%), 난각이 기형이거나 깨진 경우 11건(3.9%), 난 내용물에 혈반이나 육반등의 이물이 있는 경우는 42건(14.7%), 상한 계란은 5건(1.8%)이 유통되었다. Haugh Unit(HU)검사결과 평균은 56.1이었으며, HU 31 미만의 품질저하 계란은 18개(6.3%)이었고,총 57종 중 12종(21.1%)에서 1개 이상의 품질저하 계란이 조사되었다. 난각의 일반세균 오염수준은 <10-8.2$\times$$10^3$cfu/ml로 고르게 분포하였고, 대장균은 8건(4.7%)이 검출되었다. 총 28개의 계란(난황)에서 Salmonella enteritidis는 검출되지 않았고, 계란에 잔류하는 항생물질은 1종에서 테트라사이클린 계열이 검출되었다. 그 동안 위생관리의 사각지대에 있었던 계란에 대한 품질관리와 유통개선을 위한 법적, 제도적 근거가 마련되면, 계란의 표시 및 품질기준 등 사양관리에서부터 유통단계에 이르는 전반적인 안전성 관리방안이 검토되고 보완되어야할 것이다.

• Journal of Biosystems Engineering
• /
• 제33권5호
• /
• pp.348-354
• /
• 2008

Egg grading is determined by exterior and interior quality. Among the evaluation methods for the egg quality, a candling method is common to identify eggs with cracked shells and interior defects. But this method is time-consuming and laborious. In addition, practically, it is challenging to detect hairline and micro cracks. In this study, an on-line inspection system based on acoustic resonance frequency analysis was developed to detect hairline cracks on eggshells. A roller conveyor was used to transfer eggs along one lane to the impact position where each of eggs rotated by the roller was excited with an impact device at four different locations on the eggshell equator. The impact device was consisted of a plastic hammer and a rotary solenoid. The acoustic response of the egg to the impact was measured with a small condenser microphone at the same position as the impact device was installed. Two acoustic parameters, correlation coefficient for normalized power spectra and standard deviation of peak resonant frequencies, were used to detect cracked eggs. Intact eggs showed relatively high correlations among the four normalized power spectra and low standard deviations of the four peak resonant frequencies. On the other hand, cracked eggs showed low correlations and high standard deviations as compared to the intact. This method allowed a crack detection rate of 97.6%.