Adhi Harmoko Saputro

Myocardial motion analysis using optical flow and wavelet decomposition

The abnormalities of myocardial can be predicted quantitatively by analysing visualization of motion estimation in all or part of myocardial segment. To visualize the heart motion on myocardial, ultrasound image is evaluated globally using a differential optical flow technique that recursively applied a set of multi scale ultrasound image. A set of multi scale images is generated using approximation channel of Haar wavelet decomposition. The method was validated and evaluated on the real ultrasound image sequences that were collected from several health and unhealthy volunteers based on the American Heart Association standardization. The method was also compared to other methods using the same of ultrasound images. The result using our proposed algorithm is more accurate than the other methods. The results of myocardial motion estimation are shown in needle view of motion field in all myocardial segments and according to the expert echocardiographic analyzing.

Automatic tilting correction system for inhibition zones dimension measurement using low-cost camera

Currently, manual tools such as rulers or calipers have been used to measure the diameter of the inhibition zone of an antimicrobial. The inhibition zone is the main indicator of an antimicrobial effectiveness. The current camera technology supports for a virtual instrumentation application in measuring the dimension of an object. The presence of another supporting device such as light sources, causing the camera is not an upright position to the measured object. In this study, an automatic tilting correction system is introduced using the low-cost camera to measure the dimension of the inhibition zones in the flexible tilting angle. The hardware of the system consists of a set of the tilt measuring instruments, a source light and a camera that connected to PC. The software consists of image processing techniques performing camera calibration, spatial calibration, geometrical transformations and transforming pixel value to a meter value. The experiments have been conducted using standard checkerboard calibrator objects and inhibition zone objects at various orientations and the angle of rotation of the object. Tilt correction system improves the measurement system performance with a reduction in estimated errors up to 39,6%. The maximum angle with a tolerance level that is acceptable is 40°. The system has been successfully implemented and could be used as an instrument for antimicrobial effectiveness evaluation in flexible tilting position.

Antimicrobial effectiveness measurement using non-metric camera

Currently, an antimicrobial effectiveness that indicated by the diameter of the inhibition zones of antimicrobial was measured using manual tools such as rulers or calipers. In this paper, a camera-based system for virtual measurement of the diameter of the inhibition zones is introduced using a non-metric camera. The hardware consists of a set of measuring instruments, light source, and a camera that connected to PC directly. The software consists of image processing techniques performing camera calibration, spatial calibration, virtual ruler interface and transforming pixel value to a meter value. A standard checkerboard with certain dimension and shape was used as calibrator object to generate camera and spatial calibration parameter. The measurement system was evaluated using a set of standard checkerboard object at various camera object distances, camera angles and object orientation. The average measurement error between actual and computed inhibition diameter obtained from the antimicrobial object are 0.932%, 0.847% and 1.136% at camera-object distance 15 cm, 20 cm and 25 cm, respectively. Finally, the proposed measurement system is very promising and could be used as an instrument for antimicrobial effectiveness evaluation.

A non-destruction measurement system based on hyperspectral imaging for sugar content in banana (Musa sp.)

Sugar content is one of the important parameters to determine the quality of banana. In this paper, a non-destruction measurement system for sugar content is introduced using hyperspectral camera system over the VIS/NIR (400–1,000 nm) spectral range. Hyperspectral image (HSI) calibration was performed to compute reflectance value of banana surface in full wavelength range while spectral and spatial analysis was conducted using a partial least squares regression (PLSR) to create a model that computing relationship between the HSI spectra and the sugar content. The ground truth value of sugar content was measured using digital refractometer on the extracted banana sample. The proposed system was evaluated using 90 Ambon bananas (Musa acuminata Colla) which consist of 30 raw, 30 mature and 30 overripe banana. The PLSR model provided the root mean square error of 0.79 % and the correlation coefficient R2 of 0.988 in the full wavelength band. Finally, the proposed non-destruction prediction system could be implemented as an instrument for sugar content measurement of banana fruit.

Antimicrobial effectiveness measurement system based on circle detection algorithm

Currently, antimicrobial effectiveness parameter is predicted by measuring the inhibition zone diameter manually. In this paper, an antimicrobial effectiveness measurement system was introduced using circular Hough transformation method. The hardware of measurement system consists of a set of measuring workbench, two halogen light sources and a camera that connected to PC via USB. The software of system consists of image acquisition, camera calibration, pre-processing, Hough transforms and meter conversion. A checkerboard with known dimension and shape was used to calibrate the camera and to generate spatial calibration parameter. A contrast enhancement was performed by adding object background using strong color. The morphological method was used to detect the edge of the inhibition zone boundary. Then, the edge was used as an initial point to construct the circular shape using Hough transformation method. An antimicrobial effectiveness was measured based on the diameter alteration of inhibition zone which formed by an antimicrobial in agar subtract. The average measurement error of inhibition zone obtained from the antimicrobial object was 1.05% and 1.09% at camera-object distance 12.2 cm and 17.2 cm, respectively. Finally, the proposed antimicrobial effectiveness measurement system could be used as an instrument for antimicrobial effectiveness testing.

Carotenoid profile map system based on hyperspectral technique in banana (Musa sp.)

Commonly, the total carotenoids content in fruit is measured using spectrophotometric analysis based on color absorption levels. In this paper, a bananas carotenoids profile map system was introduced using Hyperspectral imaging technique combining spectral and spatial analysis. The profile map provided the distribution of total carotenoids on the surface of the banana. The systems consist of the measurement system and the set of a mathematical model which transforms each pixel of spatial to a value representing the number of the carotenoids in banana. The measurement system composed a workbench, halogen light source, and hyperspectral camera that connected to PC using Camera Link interfaces. The partial least square regression was used to compute transformation model between the reflectance and the total carotenoids value. The ground truth value of carotenoids content was calculated on banana peel using Sims-Gamon methods. The proposed system was evaluated using 30 samples of Ambon bananas (Musa paradisiaca) which consist of 10 samples for each maturity level. In the full wavelength range, the PLSR model provided the root mean square error of 0.7934 % and the correlation coefficient R2 of 0.94 in predicting the total carotenoids content in banana. Finally, the proposed system could be used in analyzing the distribution of the carotene in banana.

Bananas moisture content prediction system using Visual-NIR imaging

Commonly, the fruit moisture content could be measured by comparing the mass decrement of an object through the oven drying method. Bananas moisture content prediction system was introduced using the Visual-NIR imaging technique. The hardware of the proposed system consists of a set of the workbench, controllable slider, two halogen light sources and a hyperspectral camera that connected to PC via Camera Link. The software of system consists of reflectance image profile measurement, feature extraction on spectral and spatial data, and moisture prediction model. The reflectance image profile was extracted from the banana surface based on current image, white and dark image reference. The feature sets were computed using a principal component analysis (PCA) and partial least square regression (PLSR) on the full wavelength range of HSI spectra. The purpose of using two regression methods in this research is for comparing the result of moisture content prediction. The proposed system was evaluated using 45 Raja bananas (Musa textilla) samples which consist of 15 samples for each maturity stage. The prediction error between predicted and measured data with PCR is 0.58 % and produce correlation coefficient R2 of 0.79. The PLSR model of banana content prediction system has RMSE 0.25% and R2 0.96. The results show that the proposed system can predict the banana moisture content and suitable in an industrial sorting system for banana fruit quality.

Local and global human activity detection for room energy saving model

Energy saving is an effort to decrease and minimize unnecessary energy consumption. The energy saving and energy efficiency is one of the famous issues in the last decade since the energy resource is rapidly depleted. Reducing unnecessary energy consumption could be performed by architectural design or automatic system approach. This paper proposes an idea to develop a model of energy usage in a room by detecting and counting the number of peoples and detecting their activities. In order to achieve the energy consumption model, the method consists of two modules: the human detection and the activity quantization of each detected object. The human is detected using multiple registered image channels that computed in linear and non-linear transformations of the input image. The activity of each detected object is approached by computing the movement of an object in a certain time. Moreover, the human activities were represented based on local and global motion of all detected object. Videos were recorded in a classroom to evaluate the performance of the method. In evaluation, the cooling load weight model of student walking and pointing hand activities were computed in certain schema. The results show that the method can predict human activity effectively. The proposed technique is suitable for a small cost of the room energy controller and achieves the goal of energy saving.

GLOBAL FEATURE FOR LEFT VENTRICULAR DYSFUNCTION DETECTION BASED ON SHAPE DEFORMATION TRACKING

Left ventricular (LV) shape alteration is closely correlated with cardiac disease and LV function. In this paper, we propose a feature to detect LV dysfunction globally by analyzing the LV shape deformation in systolic contraction. The feature is an index that is extracted from geometric measurement of LV shape such as the length of the long axis, the short axis, and the apical diameter. A framework for computing the features is also proposed that consists of shape model construction and motion estimation of myocardial boundary. The LV shape model is extracted from apical 2 and 4 chamber views of 2D echocardiography. The long axis, the short axis, and the apical diameter were redefined according to the LV shape constructed. An optical flow technique was used to estimate the position of the LV boundary in each frame. The classification of the LV dysfunction was performed using linear discriminant analysis (LDA) and neural networks (NNs). The 2D echocardiography dataset collected from routine clinical check-up were used to validate the proposed method by comparing the computation result and cardiac expert diagnose. Classification performance and statistical analysis, which was performed to discriminate between healthy and diseased data, indicated promising results. The global LV features would provide a strong basis for a global LV function diagnosis and a global cardiac pathology assessment.

Shape deformation descriptor using Fourier analysis

Deformation analysis of left ventricle (LV) shape could provide a new quantitative understanding of its abnormality. Currently, there is established motion estimation that allows accurate tracking of every point on the 2D echocardiography (2DE). This method produces a precise movement vector of each point in 2DE sequence. Analyzing this data using Fourier analysis could produce a new pattern to determine normal and abnormal deformation of LV. Observation of this method was performed on dataset acquired from 10 normal subjects and 10 patients. Two standard views (apical 2 and 4 chamber) were analyzed using a proposed technique to determine a novel insight of deformation. The results obtained are very promising and could be used as reference for future cardiac abnormality detection.