Classification of Tomato Maturity Level with Light Reflection Method using LDR Sensor Muhammad Hafizhankarim Wachid (a), Lusianti Surya Ningsih (a), R. Silvia Putri Raharja Effendi (a), Dian Ahmad Hapidin (a,b), Khairurrijal Khairurrijal (a,b)
a)Department of physics, Institut Teknologi Bandung, Jalan Ganesha 10, Bandung 40132, Indonesia
b) Research Center for Bioscience and Biotechnology, Institut Teknologi Bandung, Jalan Ganesha 10, Bandung 40132, Indonesia
Abstract
Tomato is one of the most popular plants in home-scale plantations because it is relatively easy to grow especially in tropical climates. Tomatoes production and distribution are high but the process of maturity is quick. Farmers require to classify the level of tomato maturity accurately in order to reduce the risk of decay during the distribution and storage process. Many tomato maturity classification methods utilize reflected light such as Linear Discriminant Analysis (LDA), Color Blob, Learning Vector, and Artificial Neural Network (ANN). However, these methods have disadvantages in cost, complicated data processing, and time of processing data. Therefore, a low cost and reliable system for classifying the tomato maturity is needed. This paper discusses a system to classify tomato maturity with light reflection using light dependence resistor (LDR) sensor. The tomato maturity classification was based on reflected light received by the LDR with the light source was from red-green-blue (RGB) LED. LDR and LED were covered with a non-transparent black container to isolate the sensor from ambient light. The light emitted from the LED would be reflected by the tomato skin, then the reflected light was captured by the LDR. LDR was connected to a circuit containing a Wheatstone bridge and a differential amplifier with an output connected to Arduino module. The LDR resistance will change followed by the change in the output voltage of the Wheatstone bridge, which will be amplified by the differential amplifier. In this study, we used the method of comparison of experimental results with the results of reference color measurements that printed on paper. This method is able to classify the maturity of tomatoes into 6 level based on United States Standards for Grades of Fresh Tomatoes. There are red, light red, pink, turning, breakers, and green. Experimental data and reference data were obtained with the same pattern for each level of maturity. This device has a high precision with output standard deviation of 0.0786 V for red reflection, 0.0375 V for green reflection, and 0.0033 V for blue reflection. A light reflecting method using LDR sensor is good enough for the classification of tomato maturity level. The instrument design allowed the measurement result unaffected by the intensity of ambient light so that higher accuracy and precision can be achieved.
