Abnormal erythrocytes have diverse shapes. The appearance of specific erythrocyte shapes in a person’s blood can indicate certain diseases, including thalassemia. We used thalassemia peripheral blood smear images and applied a segmentation process to produce single erythrocyte sub-images. Each erythrocyte has a unique shape. The selection of appropriate features to represent erythrocytes is critical for classification accuracy. We used morphological features such as moment invariants, geometry parameters of the cell and central pallor, and distance angle signature (DAS) morphological features of the cell and central pallor. We combined morphological features with texture and color features to increase the accuracy of erythrocyte classification. In this study, the multi-layer perceptron is used to classify nine shapes of erythrocytes present in thalassemia cases. The experimental results of 7108 erythrocytes indicated an accuracy of 98.11% based on the combination of features. The experimental results also show that the combination of features we proposed produced higher classification accuracy than previous work, which yielded an accuracy of 93.77%.

Hematological disorders of leukocytes (white blood corpuscles) are very frequent in medical practices. Each year, the number of new cases of leukemia increased with a high mortality rate what can cause a delay in diagnosis and treatment of leukemia. The hematologist faces difficulty in classifying white blood cells included cancerous leukemia cells. Currently, the diagnosis of hematological disorder was made by microscopic analysis of peripheral blood smear by hematological experts. However, this kind of diagnosis is complex, costly process, time-consuming and inherently subjective, hence to avoid such problems a classification technique based on computer vision is required. This document provides a survey of many kinds of literature which talk about leukemia detection using image processing. In this year, many improvements were made in this field. With this document, we provide a new perspective from many kinds of literature so future research can be developed effectively for their research.

An electronic nose (e-nose) with twelve MOS-based gas sensors have been used to classify the quality of the Indonesia black tea. The black tea samples were obtained from local tea shops in Yogyakarta, including first quality (Q1) or BOP (Broken Orange Pokoe), second quality (Q2) or BPII (Broken Pokoe II) and third quality (Q3) or Bohea. For each dry tea sample was tested 31 times, at temperatures (40.0 ± 1.0) °C without any treatment and it analyzed by unsupervised principal component analysis (PCA). The experimental results showed that all three samples almost have the same aroma when observed from the sensor response. The difference from the aroma samples lie on the sensor response of the signal intensity and it can be classified clearly by the PCA with first and second principal component accounted for 80.3% and 15.3% of the variance, respectively. Besides that, the analysis of PCA indicates that the sample Q2 is closed to Q1 and Q3. It suggests that a small constituent element of black tea in Q1 and Q3 also contained in Q2. The overall results e-nose with the non-selective gas sensor can be used to classify the quality of Indonesia black tea properly.

A Fuzzy Based Decision Support System for Evaluating Land Suitability and Selecting Crops

Implementation of automation technology for grading tobacco leaf was very promising. In Indonesia, grading tobacco leaf was done manually and relied on the skill and experience of tobacco leaf graders. Large tobacco plantation needed many graders, and the workers needed to be trained, to become a skilled grader. It would take a long time and substantial cost to prepare sufficient graders. Even if the plantation had enough graders, monotonous and long duration of work would raise the human error. Therefore, we proposed a method for grading tobacco leaf based on color and quality using image processing techniques. This work covered quality inspection of tobacco leaf, namely leaf defect detection and classification of tobacco leaf based on color. Image processing techniques such as image thresholding, morphological operation, blob detection, and color analysis of tobacco leaf were employed to determine the grade of tobacco leaf. From the experiment, the proposed method was able to detect a leaf defect and able to classify tobacco leaf with 91.667% accuracy.