Yeni Herdiyeni

I-PEDIA: Mobile application for paddy disease identification using fuzzy entropy and probabilistic neural network

This research developed a mobile application for paddy plant disease identification system using fuzzy entropy and classifier probabilistic neural network (PNN) that runs on Android mobiles operating system. Paddy diseases are a major cause of yield loss and lower profit in rice production. Paddy diseases are extracted from digital paddy leaf images using fuzzy entropy and then the diseases are classified using PNN. Cross validation is used for assessing how the results of a statistical analysis will generallize to an independent data set. The experiment result shows that the accuracy of paddy diseases identification is 91.46%.

Fusion of Local Binary Patterns features for tropical medicinal plants identification

Selection of leaf features that are appropriate for identification is very important. Local Binary Patterns (LBP) is one of texture feature that is efficiency and robustness for plant identification. Meanwhile, in LBP we have to define good size of sampling point. In this research we propose fusion of LBP features, which incorporates different size of sampling point. There are two ways for fusion of LBP. First, we perform a straightforward fusion by calculating histogram of multiple LBP features separately using varying the size of sampling points and radius, then concatenate the multiple histograms together. Second, each histogram of LBP features is classified, and the feature fusion can be accomplished by classifier combination. For leaf classification, we used probabilistic neural network (PNN) to classify LBP features. The experiment performed on tropical medicinal plants and house plants. According to experimental results, the fusion of LBP features can improve accuracy in plant identification. This system is very promising to help people identify medicinal plant automatically and for conservation and utilization of medicinal plants.

Comparison of 2D and 3D Local Binary Pattern in Lung Cancer Diagnosis

Comparative study between 2D and 3D Local Binary Patter (LBP) methods for extraction from Computed Tomography (CT) imagery data in lung cancer diagnosis is conducted. The lung image classification is performed using probabilistic neural network (PNN) with histogram similarity as distance measure. The technique is evaluated on a set of CT lung images from Japan Society of Computer Aided Diagnosis of Medical Images. Experimental results show that 3D LBP has superior performance in accuracy compare to 2D LBP. The 2D LBP and 3D LBP achieved a classification accuracy of 43% and 78% respectively.

Leaf Shape Recognition using Centroid Contour Distance

This research recognizes the leaf shape using Centroid Contour Distance (CCD) as shape descriptor. CCD is an algorithm of shape representation contour-based approach which only exploits boundary information. CCD calculates the distance between the midpoint and the points on the edge corresponding to interval angle. Leaf shapes that included in this study are ellips, cordate, ovate, and lanceolate. We analyzed 200 leaf images of tropical plant. Each class consists of 50 images. The best accuracy is obtained by 96.67%. We used Probabilistic Neural Network to classify the leaf shape. Experimental results demonstrated the effectiveness of the proposed approach for shape recognition with high accuracy.

Framework model of sustainable supply chain risk for dairy agroindustry based on knowledge base

The objective of this paper was to develop a framework model for sustainable supply chain risk for dairy industry based on knowledge base. It presented a conceptual framework with integrated risk supply chain and knowledge base systems. The critical point of dairy located on the product which has the characteristic easy damage. Risk-damaged dairy contaminated with bacteria due to improper handling of dairy. Risk occurred in each activity in the supply chain network ranging from farmer, cooperative and dairy processing industry. The structured approach of supply chain risk divided into the phases of risk identification, risk measurement and risk assessment, risk evaluation and risk mitigation and contingency plans; and risk control and monitoring system based on knowledge base system. Adding Knowledge base component to risk supply chain will produce the following process: knowledge base risk capture, knowledge base risk discovery, knowledge base risk examination, knowledge base risk sharing, knowledge base risk evaluation and knowledge base risk repository. The relationship between risk factor, risks and their consequences are represented on Failure Mode and Effect Analysis (FMEA) and Hierarchical Risk Breakdown Structure (HRBS). Likelihood of risk event occurring, the level of dependence between risks and severity of risk event are quantified using linguistic variables and fuzzy logic. The proposed system was designed by Intelligent Decision Support System (IDSS). The design of this model was able to improve the effectiveness of decision-making with regard to the organization of knowledge, storage and sharing of knowledge in the agro-industry supply chain risks dairy.

SADE: Android spectral reflectance estimator application using Wiener estimation to estimate sambiloto leaf's age

This research proposes an Android application to estimate sambiloto (Andrographis paniculata) leafs age from its estimated spectral reflectance using Wiener estimation. Sambiloto is one of Indonesias popular medicinal plant. In order to use quality plants, a quality control method, such as lab tests, must be conducted. These lab tests require the destruction of leaf samples. One promising alternative is by using image processing using Wiener estimation. Wiener estimation is a conventional method to estimate high-dimensional data from low-dimensional data, for example in this case, three-channel image (RGB) to spectral reflectance. We can quantify the sambiloto leafs quality through its spectral data in the form of its age. This research also proposes an improvement in dataset acquisition for the Wiener estimation. In the experiment we used datasets consisting of 97 standard colors, 15 samboloto leaves, and their combination. The results shows that the 15 sambiloto leaves dataset and second polynomial order gives the best reconstructed spectral reflectance. The RMSE and GFC of this dataset are 3.57 and 0.99, which is better than several previous researches. We use Probabilistic Neural Network for classifying the leafs age from its reconstructed spectral reflectance. The accuracy for the age identification using PNN is 65%.

Leaf vein segmentation of medicinal plant using Hessian matrix

This paper proposes a leaf vein segmentation using Hessian matrix. Leaf venation pattern is a biometric feature that form the basis of leaf characterization and classification. It is specific in certain species thus it can be used as a key feature. Hessian Matrix is a method of the second derivative ridge detection that can be used to segment the image based on its group structure by analyzing eigenvalues of the pixel. We applied thinning to achive the better result of leaf vein. In addition, we performed morphological image processing to fix broken ridges or unconnected leaf veins. We have evaluated four veins type of 80 digital leaf. The experimental results show that 53.75% of leaf image scored 2 and 42.5% scored 1 which means our proposed method has good performance to extract the primary, secondary veins and tertiary leaf vein. This method is promising to help botanist and taxonomist identifying medicinal plant species automatically.

Multiscale fractal dimension modelling on leaf venation topology pattern of Indonesian medicinal plants

This research proposed a new model to differentiate leaf venation topology patterns using Multiscale Fractal Dimension. Identification of medicinal plants is important considering wide range of biodiversity in Indonesia and significant role of medicinal plants in Indonesia. Plants identification can be performed with shape analysis using plant leaf venation as a feature. Multiscale Fractal Dimension is a shape analysis method that analyze shapes through its complexity. In this research three Indonesian medicinal plants species has their leaf venation topologies modelled with Multiscale Fractal Dimension. The result shows that while the difference is not remarkably clear, there are irregularities that can be made more evident with multiscale analysis. Future works can include Multiscale Fractal Dimension as one technique to identify plants.

A Bayesian network approach for image similarity

This paper proposed Bayesian Network approach for image similarity measurement based on color, shape and texture. Bayesian network model can determine dominant information of an image using occurrence probability of images characteristics. This probability is used to measure image similarity. Performance of the system is determined using recall and precision. Based on experiment, Bayesian network model can improve performance of image retrieval system. Experiment result showed that the average precision gain up of using Bayesian network model is about 8.28 %. The average precision of using Bayesian network model is better than using color, shape, or texture information individually.

Predicting the Spread of Acacia Nilotica Using Maximum Entropy Modeling

Acacia nilotica planted in Baluran National Park aims to prevent the spread of fire from savanna to teak forest became developed into invasive and led to a decrease in the quality and quantity of savannas. Therefore, it is required to predict the spread of A. nilotica to minimize the impacts of invasion on savanna area. The study aims to identify environmental factors which affect spread of A. nilotica. Furthermore, the spread of A. nilotica is predicted using Maximum Entropy. Maximum Entropy is efficient model since it uses presence-only data while the most of other models use presence and absence data. The experimental results reveal six environmental factors, including elevation, slope, NDMI, NDVI, distance from the river, and temperature were identified affecting the spread of A. nilotica. The most dominant environmental factors were elevation and temperature with 40% and 39.6% contributions. Maximum Entropy performed well in predicting the spread of A. nilotica, it was indicated by AUC value of 0.938.