Open Access Open Access  Restricted Access Subscription or Fee Access

Computer Vision for Black Tea Grading Based on the Black Tea Particles and Tea Infusion Color

Suprijanto, Endang Juliastuti, Amalia Rakhmawati


Black tea has been one of the main Indonesian export commodities. Since the quality of the tea determines its value on the market, therefore quality assessment is needed to warrant that black tea production meet a certain quality standard. In Indonesia, the black tea standard grading refers to Indonesian National Standard (SNI) of black tea 01-1902-1995. At present, organoleptic evaluation as visual and aroma inspection by trained evaluator, chemical instrument as gas chromatography and colorimetric evaluation, are being used as the tea grading. However, this method is time consuming, laborious, expensive and sometimes inaccurate. This research was intended to develop compact computer vision prototype for grading Crushing Tearing Curling – Broken Pekoe1- black tea based on particle morphology and the infusion color. In image acquisition process, standard box with illuminator and color digital microscope were used to capture the details of tea particles and tea color. On each sequence of particle morphology evaluation, 2,84 gram tea that approximately consists of 1000 granular particles was used. Morphology operation such as the parameter of area, perimeter and bending energy were applied to label each particle. The color information of tea particles and tea infusion were extracted by use of image color histogram on red and blue channels. Each parameter value was represented in a certain interval of histogram and uniformly applied in all feature sets. Multilayer Perceptron-Artificial neural network (MLP-ANN) was used to classify the tea grade on class A,B and C. Two architecture MLP was trained with input of 30 data sets of tea particles and tea infusion color with A, B and C grade using backpropagation. The ANN was validated by use of 30 datasets (A, B, C grade) apart from the training data. The validation process yielded 100% result in recognizing the black tea grading which matched the organoleptic method.


Black tea, grading, particles, infusion, computer vision, morphology, geometry feature, color, artificial neural networks

Full Text:



  • There are currently no refbacks.

Disclaimer/Regarding indexing issue:

We have provided the online access of all issues and papers to the indexing agencies (as given on journal web site). It’s depend on indexing agencies when, how and what manner they can index or not. Hence, we like to inform that on the basis of earlier indexing, we can’t predict the today or future indexing policy of third party (i.e. indexing agencies) as they have right to discontinue any journal at any time without prior information to the journal. So, please neither sends any question nor expects any answer from us on the behalf of third party i.e. indexing agencies.Hence, we will not issue any certificate or letter for indexing issue. Our role is just to provide the online access to them. So we do properly this and one can visit indexing agencies website to get the authentic information.