Endang Juliastuti

Quantitative image analysis of periapical dental radiography for dental condition diagnosis

Periapical dental radiography have been used by the dentist to diagnose the lessions of the tooth. In Indonesia, recent development of this dental technology still used conventional method by using negative film with several limitation. This research was conducted to study digitalization of periapical film using digital dental X Ray reader for dental condition diagnosis. This dental condition was made by two step image analysis, consist of qualitative analysis and quantitative analysis the dentin and pulp. To segmen dentin and pulp, active contour was used to simplify the image analysis process. Qualitative analysis was made by the help of the visual inspection of the dentist, while quantitative analysis was made by compute various statistic parameter. Result of this research show that varians and the intensity ratio between dentin and pulp is good enough to be statistic parameter to differentiate the condition of the dental.

Ultrasonic tomography for reinforced concrete inspection using algebraic reconstruction technique with Iterative Kaczmarz method

Non-destructive testing for evaluating the safety and quality of concrete structures is highly needed in the field of construction and civil engineering. The imaging of concrete structures is still, however, considered to be a very challenging task due to the non-homogeneous properties of the material. In order to address this challenge, this paper presents a non-destructive testing method using ultrasonic wave to identify the internal structure of a concrete sample. A combination of signal smoothing and threshold method was used as signal processing tool for interpreting and analyzing the ultrasonic waveform from the concrete sample. The Algebraic Reconstruction Technique with Iterative Kaczmarz method was then applied as the image reconstruction algorithm. The results from our experiment show that the ultrasonic tomography based on the Iterative Kaczmarz method successfully distinguish embedded steel rod from the rest of materials in the concrete. This paper thus illustrate the proven reliable of the proposed method for testing concrete structures.

Microscopic Surface Measurement Using Phase Shifting Method

Accurate information of microscopic topography is very important for efficacy assessment of a surface texture of skin health. Due to the limitations of the direct visual assessment of skin microscopic topography, an optical dermastocopy is very common to be used as skin imaging device to magnify skin topography based on a white light reflection. The limitation of this method is its poor spatial resolution to quantify skin topography. In this work, microscopic skin imaging based on phase shifting method is configured using a DLP pico-projector with LED illumination and a handheld digital microscope. As illuminator for the digital microscope, the DLP projector is programmed to generate patterned light on skin surface. Image processing is required in providing accurate information of surface topography. The first step, a wrapped phase shifting must be extracted from acquired intensity images. The second step is obtaining unwrapped phase image, which is a critical process because it must be recovered from wrapped phase shifting that containednoise. Finally, phase offset due to multiples of 2π during phase unwrapping must be removed. Early experiments on simple object are carried out to test the level of distortion of fringe in several variations of contrast and also to test the performance of the system on several frequency variations. The test results indicate the depth proportion obtained from absolute phase image has the same trend as the proportion of direct measurement. Implementation on the skin surface profile performed on three test areas: the back of the hand and knuckle creases. Based on quantitative and qualitative analysis,our proposed scheme of skin imaging based on phase shifting is promising for surface profile measurement and imaging of the skin.

Image reconstruction of time domain diffuse optical tomography for quality control on seed potatoes

The purpose of this study is to apply model based iterative reconstruction method on a ring array diffuse optical tomography system based on mean time of flight. This study was conducted on numerical object that represents the real condition of seed potatoes. The object was illuminated by the near infrared source from 12 positions on objects boundary. A set of near infrared detector are placed on the periphery of the object and it measures the intensity of propagated light. In the simulation, we vary the condition of object then we analyze the correlation between simulated and reconstructed image. The result of this study indicated that time domain-diffuse optical tomography is promising for quality control on seed potatoes.

Dental panoramic image analysis on mandibular bone for osteoporosis early detection

Osteoporosis is a degenerative disease characterized by low bone density and micro architectural deterioration of bone tissue with a consequent increase in bone fragility and decreasing bone mechanical force on supporting body normal activity. One of common technique used for measurement bone mass, bone mineral density or other aspect related bone structure is Dual Energy X-ray Absorptiometry (DXA). Previous researchers shown opportunity to utilize dental panoramic images for early detection and estimate the probability of having osteoporosis. However, a robust image quantitative is still challenge. In the paper, quantitative of dental panoramic is reported based on Gray Level Co-occurrence Matrix (GLCM). Feature extraction from GLCM will be use as an input for Support Vector Machine (SVM) algorithm to classify normal and osteoporosis. The classification result will validate using BMD data of 23 samples prepared by Dental Radiographs Department, where panoramic images are imaged from patient postmenopausal, with ages 52-73 year. Classification using SVM with kernel function multilayer perceptron for normal and osteoporosis showed that the best performance (using 9 training data and 14 test data) was 85,71% accuracy, 90,91% sensitivity, and 66,67% specificity. Its best performance result is obtained by using contrast, correlation, energy, and homogeneity combination for SVM classification input.

Static evaluation of one shot 3D surface imaging using digital colored fringe projection technique

This research developed an alternative to grayscale Phase Shifting Interferometry for 3D surface measurement by utilizing three channels color fringe. As it has been known, grayscale PSI has limitation on dynamic object measurement due to its multi-frame term. By utilizing digital color image, it merely needs a single fringe projection making it possible to obtain faster data acquisition. However, each channel of the recorded image has different intensity range due to optical systems non-linearity that would significantly affect the quality of the reconstruction. Therefore, normalization is urgently needed before performing phase computation. In this study, the normalization was performed using Isotropic N-Dimensional Fringe Normalization (INFPN) based on Hilbert Transformation, while the phase processing includes extraction using 3-step PSI and global phase unwrapping. Experiments are conducted to evaluate RGB fringe performance compared to the grayscale 3-step PSI and grayscale single frame Fourier Transform Profilometry (FTP). The results show that RGB fringe can produce a fairly good reconstruction with the measurement error of less than 6% compared to the grayscale 3-step PSI on static evaluation. It also gives far better quality of reconstruction than the grayscale single frame FTP.

Detection of water content in lubricating oil using ultrasonics

The presence of water in lubricating oil is destructive even in very small quantities. The ultrasonic measurement is an alternative method to find a solutions for detecting and measuring the water contamination in the lubricant oil. The amount of water contamination in lubricating oil will affect the velocity of wave propagation in the lubricant. In this study, a 2.25 MHz ultrasonic wave is propagated to a contaminated lubricating oil medium. The ultrasonic wave propagation time from the transmitter to the receiver is measured so that the speed of soundwave can be obtained. Here, an artificial neural network (ANN) is employed to interpret the speed of sound into the water contamination in the lubricant. The measurement methods discussed in this paper give an average error of approximately 11.3%.

Asymmetric flow velocity profile measurement using multipath ultrasonic meter with neural network technique

Average flow velocity in a multipath transit time Ultrasonic Meter (USM) is calculated by integrating the mean flow velocities along all of the acoustic paths. Every path has a weight that represents its contribution to the average velocity. In the conventional weighting methods, the flow for the entire USM is assumed to be fully developed and symmetrical, so that each path has a fixed weight of USMs measuring range. However, if the flow is not in the ideal conditions or distorted, implementing such conventional techniques will produce measurement results with poor accuracy. To handle that problem, an advanced weighting method is thus needed, so that a set of weights of all paths is able to adapt to several circumstances. In this paper, the performance of a conventional weighting method represented by the Covered Area (CA) is compared toan advanced weighting method represented by the Artificial Neural Network (ANN) using Root Mean Square Error (RMSE) analysis. In order to find the best architecture, several ANN parameters including learning rate, number of hidden layer, and neuron, are varied. The position and number of acoustic path are also varied based on the different number of USM Tomographic transducers from 10 to 16. The numerical simulation results show that the smallest RMSE for ANN with the Cascade algorithm is 0.0065 for USM Tomo 16 trans 6 paths, while in the case of CA is 2.2 10−3 for USM Tomo 12 trans 4 path. Furthermore, the sensitivity level of ANN-Cascade to the changes in number and path position is much lower than that of CA, whereas the range of ANN and CA RMSE are (6.5 10−3∼1.5 10−2) and (2.2 10−3 ∼7.6 10−2), respectively. By using tolerance within± 1% (AGA-9), the average RMSE for the testing sets is 9.1 10−3, while the ANN-Cascade and CA are 1 10−2 and 1.8 10−2.

Static detection of flat head railways depletion using analysis of laser area and position on rail type R-54

Load and age of rails can result in problems such as breakage, depletion, and expansion that can lead to accidents. Rail inspection has been done manually by operator tracing the rails by walking or riding a special inspection vehicle. These methods obviously are inefficient and inaccurate, as operators might be missing some of the defects. In this research depletion detection of rails are conducted by analyzing changes of the area as well as position shifting of laser spot on captured images by utilizing the triangulation principle. Accuracy and efficiency improvement of rail inspection are expected from this method. Prior calibration of the system was conducted using gauge blocks with thickness varying from 19 to 1 mm with 1 mm decrement. Area changes and position shifting of laser spot are later analyzed through image processing. The system was also implemented on R-54 rail type based on the calibration and later be compared to the manual measurement data. It was shown that the system can detect depletion in rail type R-54. The calibration result shows that the deviation percentage of the measurement of laser area are ranging from 11.41% to 13.48% while for the measurement of laser spot position shift is from 6.91% to 8.80%. Implementation on rail type R-54 shows the deviation percentages between proposed method and manual measurement are ranging from 1.52% to 10.04% for the area measurement, while for the position shifting ranged from 1.11% to 12.68%.

Dental panoramic image analysis for enhancement biomarker of mandibular condyle for osteoporosis early detection

Osteoporosis is a degenerative disease characterized by low Bone Mineral Density (BMD). Currently, a BMD level is determined by Dual Energy X-ray Absorptiometry (DXA) at the lumbar vertebrae and femur. Previous studies reported that dental panoramic radiography image has potential information for early osteoporosis detection. This work reported alternative scheme, that consists of the determination of the Region of Interest (ROI) the condyle mandibular in the image as biomarker and feature extraction from ROI and classification of bone conditions. The minimum value of intensity in the cavity area is used to compensate an offset on the ROI. For feature extraction, the fraction of intensity values in the ROI that represent high bone density and the ROI total area is perfomed. The classification will be evaluated from the ability of each feature and its combinations for the BMD detection in 2 classes (normal and abnormal), with the artificial neural network method. The evaluation system used 105 panoramic image data from menopause women which consist of 36 training data and 69 test data that were divided into 2 classes. The 2 classes of classification obtained 88.0% accuracy rate and 88.0% sensitivity rate.