Classification of the Quality of Milk Using Spectrophotometer System Based on Raspberry Pi

Dairy cow milk is a beverage that is often consumed by many people. The quality of milk for dairy cows needs to be maintained because milk naturally experiences relatively rapid decay. Protein content can be used as a parameter used to determine the quality of milk. One method used to analyze protein content of milk is by using a spectrophotometer. Technically, a spectrophotometer in a laboratory serves to measure the transmittance or absorbance of a sample expressed in a wavelength function. The wavelength of a sample varies according to the color of the spectrum used. To test protein content, a biuret spectrophotometry method is used which produces purple in the sample, so that the wavelength with the highest absorbance value is found in the range of 525 nm. The absorbance value of the tested samples of dairy cow milk shows the quantity of protein content. The spectrophotometer system is designed using image processing based methods equipped with Raspberry Pi. Tests were carried out by comparing protein content between standard laboratory spectrophotometers and proposed Raspberry Pi based spectrophotometers. The comparison results obtained from Raspberry Pi based spectrophotometer tests have an average error of 2.366%. The classification of the quality of milk for dairy cows produces 3 levels of classification, namely milk for dairy cows with a protein content of 0% - 2.5% indicating low quality, 2.51% - 7.38% indicating medium quality and 7.39% - 9.63% indicating high quality.