Balza Achmad

Inter-frame Enhancement of Ultrasound Images Using Optical Flow

Ultrasound imaging, also known as ultrasound scanning or sonography, is a very popular medical test that helps medical doctors diagnose and treat medical conditions of their patients. However, one common problem that persists is that ultrasound images suffer from speckle noise that degrades their quality. In this paper we present an enhancement technique for ultrasound images by making use of three consecutive frames extracted from an ultrasound video. The technique uses the optical flow algorithm to reconstruct an intermediate frame based on the preceding and the following frames. The reconstructed image is then utilized to enhance the middle frame by mean of fusion. Based on the test, the best result is achieved using Lukas-Kanade optical flow and average operator.

Visual-based fuzzy navigation system for mobile robot: Wall and corridor follower

This paper presents the development of a visual-based fuzzy navigation system that enables a mobile robot in moving through a corridor or following a wall. The system employs a camera to detect the existence of walls on the left, the right, and the front of the robot. A mamdani-type fuzzy logic controller uses the information gathered by the camera to determine the turning angle and the speed of the robot. The fuzzy system is tested using an OpenGL-based 3D simulator that capable in animating the movement of the robot as well as generating the images captured by the camera. The results of the test confirm that the controller shows a good performance in navigating the robot.

Warped anisotropic diffusion of ultrasound image

Ultrasound images contain speckle noise that creates granular pattern which degrades their quality. Typically, the granular noise has a circular pattern that circles the position of the ultrasound probe which acts as it center. As such, anisotropic diffusion filter cannot completely remove the granular noise. In this work, we propose a technique that pre-processes an ultrasound image using warping operation prior to the application of anisotropic diffusion as well as post-processes it via dewarping operation. Based on visual observation and power signal-to-noise ratio (PSNR) calculation of the test image, the proposed warped anisotropic diffusion (WAD) technique provides an improve result when compared with the original version of anisotropic diffusion (AD) technique.

Localization of white blood cell images using Haar Cascade classifiers

Haar Cascade Classifier is a method for detection of objects within an image, which is widely applied on face detection. This paper discusses the utilization of Haar Cascade Classifier in locating the positions of white blood cells in an image. The results showed that this method is able to localize white blood cells with precision and recall values of 95% and 74% respectively. This method is also able to distinguish white blood cells from other objects that have color resembling white blood cells.

Development of Plant Growth Monitoring System Using Image Processing Techniques Based on Multiple Images

The research aimed at developing plant growth monitoring system based on image processing technique from images captured using multi-cameras (webcams). The output of the research was a monitoring system equipped with graphical user interface (GUI) to monitor the plant growth and nondestructively predict age and fresh weight of mustard (Brassica rapa var. parachinensis L.). Five webcams on five different positions were installed on the image-capturing box to capture the plant images every 3 days up to 48 days. Then, 1050 images of 70 plants on vegetative period resulted. Of these, 700 images were used for training and 350 images for validation. An artificial neural network (ANN) was used to predict age and weight of the plants. The network architecture consisted of four layers with five input neurons, first hidden layer with five neurons, second hidden layer with five neurons, and output layers with one neuron. Training function of ANN used was trainlm and learning rate value was 0.001. The activation function for predicting plant age and plant weight was logsig-logsig-tansig and logsig-tansig-tansig, respectively. ANN models could predict plant age with an R2 value of 0.96 and fresh weight with an R2 value of 0.95.

Thermal image human detection using Haar-cascade classifier

Haar-Cascade classifier method has been applied to detect the presence of a human on the thermal image. The evaluation was done on the performance of detection, represented by its precision and recall values. The thermal camera images were varied to obtain comprehensive results, which covered the distance of the object from the camera, the angle of the camera to the object, the number of objects, and the environmental conditions during image acquisition. The results showed that the greater the camera-object distance, the precision and recall of human detection results declined. Human objects would also be hard to detect if his/her pose was not facing frontally. The method was able to detect more than one human in the image with positions of in front of each other, side by side, or overlapped to one another. However, if there was any other object in the image that had characteristics similar to a human, the object would also be detected as a human being, resulting in a false detection. These other objects could be an infrared shadow formed from the reflection on glass or painted walls.

Power estimation of G.A. Siwabessy Multi-Purpose Reactor at start-up condition using artificial neural network with input variation

Thermal power of nuclear reactor needs to be carefully maintained to produce desired electrical power. While in-core measurement system has a higher safety risk, ex-core measurement has been employed to increase safety. Artificial neural network with multi-layer perceptron architecture and Bayesian regularization algorithm has been trained and tested for estimating the thermal power at G.A. Siwabessy multi-purpose reactor. Furthermore, to find out the parameters that provide the strongest influences to thermal power, variations of input were tested to the estimation system. This study found that the output from primary coolant temperature sensor was the main factor that produces the strongest effect toward thermal power of the reactor, whereas the output from pressure sensor providing the smallest effect toward the power calculation.

A visual recognition supporting tool for mapping environmental data using handheld measurement instruments

The environmental data mapping has become an important step to measure the mitigation plans from pollution impacts. It often requires an observer to measure environmental data in the field using portable measuring devices and combines them with geolocation data to produce geomapped data. One of its technical difficulties during this process is in the step of recording the data, since the observer usually has to record measurement data manually into certain program. In this paper, we developed a smartphone application that capable in reading digits on a handheld measuring device display using computer vision. The measured data is sent to a cloud-based data management system, Google Fusion Tables in order to be displayed in the form of interactive maps with web application. The template matching and pixel counting methods were used to recognize the digits of each respective digit. The variations of pre-processing techniques and template matching methods were also tested. It is shown that the combination of pixel counting and normalized correlation coefficient template matching method with binary digit and template images input resulted in maximum accuracy. From 120 sample images with varied lighting conditions, both in lab-scaled and in-field tests, the digit recognition rate has achieved 70.83% accuracy in lab-scaled and 67.92% when tested in-field. Nevertheless, digit recognition method shown unsatisfactory result which demands different digit recognition method.

Enhanced anisotropic diffusion with noise amplification suppression

In order for medical doctors to be able to effectively utilize ultrasound images to support their diagnosis, the images require to be enhanced. The noise contained in the image needs to be reduced and the edges need to be sharpened. In this paper, an enhanced technique based on anisotropic diffusion that is capable of carrying out simultaneously image smoothing and enhancement is presented. The technique (EAD) is equipped with noise amplification suppression to prevent unwanted enhancement of noise. The technique performs well for image containing noise up to 30%. The tuning parameter is simpler compared with other anisotropic diffusion enhancements.

PENENTUAN DIMENSI TERONG JEPANG (Soalnum melongena L) DENGAN TEKNIK PENGOLAHAN CITRA SECARA WAKTU NYATA

A real-time image processing technique was applied to determine the dimensions, i.e. length and width, of eggplants. Results showed that illuminations performed a significant role; the ideal illumination type for the experiment was TL lamp (fluorescent lamp), which was placed in perpendicular position toward conveyor. Meanwhile, the ideal eggplant-positions were longwise on the conveyor. Applying those kind of circumstances, a relationship between developed program and manual measurement of eggplant length and width showed linier equation, i.e. y = x and y = 0,99 x, respectively.