Browsing by Author "Kartheeswaran, T."

Now showing 1 - 3 of 3
  • No Thumbnail Available
    Item
    Detection and Classification of Diseased Tomato Leaf Using Image Processing Techniques
    (Uva Wellassa University of Sri Lanka, 2016) Mehendran, Y.; Kartheeswaran, T.; Ediriweera, E.P.S.K.
    Tomato plants are highly vulnerable to fusarium wilt, verticillium wilt, and late blight. The symptom is yellowing of the lower leaves, which gradually wilt and die. The naked eye observation of experts is the main approach adopted for detection and the identification of plant diseases. We developed a method to detect and classify damages in leaves using image processing techniques. For this experiment, images downloaded from the interne were used. The disease regions were segmented using K-Means clustering and the classification of the disease was done with Support Vector Machine (SVM) by training with the selected features from the training set of images. The initial version has three classes such as Bacterial Wilt, Early blight and Healthy tomato leaves. The accuracy level for the identification and the classification of diseases was calculated for each category separately. The accuracy of the system for the selected nine features was calculated as 76.5%. Association among the features as Contrast, Correlation, Energy, Homogeneity, Entropy, Mean, Standard deviation, Skew, and Kurtosis gave the optimum accuracy. This system with high accuracy motivates the other researchers to extend the system with added functionality, which will be a farmer friendly software solution. Keywords: HSI, K-means, Gray-level co-occurrence matrix, Support Vector Machine (SVM),
  • No Thumbnail Available
    Item
    Detection of Dhool Number in Black Tea Manufacturing with Image Processing Techniques
    (Uva Wellassa University of Sri Lanka, 2016) Saranka, S.; Kartheeswaran, T.; Wanniarachchi, D.D.C.; Wanniarachchi, W.K.I.L.
    The possibility to use digital images of tea particles as a tool to monitor fermentation of black tea processing is studied in this project. Copper green colour is the predicted colour used to measure the degree of fermentation; therefore, determining the fermentation level by observing the copper green using naked eye is error prone and affects the complete product outcome. Black tea processing takes several batches per day, and from each batch, there are three types of particles obtain after Roll breaker processes. According to the size of the particles these are named as dhool 1, dhool 2, and dhool 3. The duration of fermentation is varied by dhool number for a given batch due to varied sizes of tea particles. Therefore, it is important to identify the dhool number for a given digital image. The method used in this project is divided in to three main phases, image pre-processing, identification of the dhool number, and prediction of the fermentation level. image processing techniques are used to extract features of tea leaves and Support Vector Machine (SVM) is used as the classifier to train the system and obtain accuracy in each stage. The results indicate higher accuracy in predicting the dhool 1 which is over 77% accurate while dhools 2 and 3 indicated accuracy levels of 69% and 73% respectively. Therefore, image processing techniques can be successfully used to predict the dhool number of a given batch of tea processing. Keywords: Fermentation, Image processing, SVM
  • No Thumbnail Available
    Item
    A Novel Method to Measure The Water Content of The Leaves Using Digital Image Processing
    (Uva Wellassa University of Sri Lanka, 2018) Wagachchi, I.A.; Rassagala, R.D.K.; Abeykoon, B.B.D.S.; Kartheeswaran, T.; Jayathunga, D.P.
    Leafy product industries like Tea, Tobacco, Palmyra, Leafy vegetables, and Ayurveda productions play a significant role to uplift the Sri Lankan economy. The water content in the leaves is an essential factor for leafy productions to maintain their quality. Naked eye observation of an expert is the general method to identify the water content. The objective of this study is to introduce a novel and easy method to measure the water content of the detached plant leaves using digital image processing. As a result, a simple computational water content prediction method has been built using image processing techniques to obtain a quality output at the end of production processes. The findings of this study help to identify the water content without an expert in an efficient manner. First, the colour images were captured in a control environment, while leaves were drying and simultaneously the weight was measured traditionally to find the water loss. Features were analysed from images to find the best features, which show a better correlation with the changes of the water content in the leaves. The basic features such as homogeneity, energy, contrast, variance, mean, median, min, max, range, kurtosis, skewness, standard deviation, entropy, correlation and IQR were extracted. The best features among the selected features have been chosen through correlation test. The classification was done with the K-Nearest neighbour algorithm by training with the selected best features of the training set of images. The green matrix of the RGB image is taken for the feature extraction to get better results. Finally, a simple model was built using the significant features which have a relationship with the water content measurement. 65.3% accuracy has been achieved, and this model can be used to predict the water content of a particular green leaf through images. This model will be a turning point for measuring the water content of the leaves in the industries in an automated manner.