Febus Reidj G. Cruz

Iris Recognition using Daugman algorithm on Raspberry Pi

Biometric is an automated recognition process that measures the unique human characteristics, such as iris recognition. Of late, different kinds of iris recognition techniques are available that can be used in different platforms. Daugman algorithm is one of the iris recognition techniques that provide high percentage of accuracy. This study aims to develop a device that performs Iris Recognition using Daugman algorithm on Raspberry Pi. Through the implementation of Daugman algorithm on Raspberry Pi, more applications may come up for iris recognition. Also to make the system adapt with the limitations of Raspberry Pi and the existing application, this work created an image processing function that preprocesses the image before it is passed to the application. Based on the tests implemented, the False Acceptance Rate (FAR) and False Rejection Rate (FRR), the system is mostly to reject registered actors than to accept impostors due to the hardware limitations.

Temperature Compensated Electronic Nose for Fruit Ripeness Determination Using Component Correction Principal Component Analysis

The study integrates a component correction algorithm on the electronic nose, intended for fruit ripeness determination, to counteract the tendency of the electrochemical metal oxide sensors to drift. The drift is any unwanted deviation of the chemoreceptor response from the true value leading to erroneous and inconsistent results. The study utilizes CCPCA — Component Correction Principal Component Analysis to approximate and remove the temperature drift component. In using this algorithm, the study theorizes that the drift acquired from the simulation of temperature drift without any sample (clean air) could generalize the drift pattern or structure in sampling Lacatan aroma samples. The PCA or Principal Component Analysis is utilized for ‘dimensioanlity reduction’ and ‘data mining’ that uncovers patterns unobserved on the multidimensional data structure. The cluster Silhouette is computed to quantitatively validate the intracluster cohesion and intercluster separation. Using PCA, the electronic nose distinguished 630 Lacatan aroma samples on the PCA Loadings — PC1 (94.65%) and PC2 (3%) and significantly improved the clustering upon removal of temperature drift. The cluster Silhouette is improved from 0.7903 to 0.8571 (unripe), 0.5358 to 0.6080 (ripe), and 0.7784 to 0.8357 (overripe).

Full-wave AC-DC converter in CMOS 0.18 micron for vibration electromagnetic energy harvest

This paper presents the design and implementation of full-wave AC-DC converter in 0.18 micron CMOS technology. Five sample voltages were taken from vibration electromagnetic harvester technology (VEHT) and simulated in three operating temperatures of 0, 25, and 70 degrees Celsius. The simulation results yielded: a maximum average power efficiency of 95.02% at 0°C 95.61% at 25°C, and 91.05% at 70°C; and an average regulation percentage of 10.45% at 0°C, 24.91% at 25°C, and 10.15% at 70°C.

Raspberry Pi and IOT Based-Automated Teller Machine Security for the DSWD 4P's Biometric System Using Fingerprint Recognition with Fast- Fourier Transform Image Enhancement, Multi-Stage Minutia Extraction

Card skimming and identity theft in ATMs transactions are prevalent nowadays. The use of the common PIN system together with the use of fingerprint as ones identification, eliminating the card skimming- and theft-prone typical card when the PIN is known, strengthens the current ATM transaction process. Images from fingerprint sensors undergo a multi-step process to provide proper user identification when transacting in ATMs. The researcher came up with the idea of designing a system to eliminate possibilities of card skimming and identity theft with the use of a fingerprint identification which is unique for each person. The system uses Fast Fourier Transform as one of its image process to produce better image quality to be used for minutiae identifications. The systems simulated performance indices showed an FAR and FRR of 0% and 15.79%, respectively, and accuracy of 92.11%. The image is remotely transmitted to a distant server wirelessly, implementing a full Internet of Things providing the system a better mobility for mobile ATM Setups in far places.

Cloud-based application for rice moisture content measurement using image processing technique and perceptron neural network

This study is about the determination of moisture content of milled rice using image processing technique and perceptron neural network algorithm. The algorithm involves several inputs that produces an output which is the moisture content of the milled rice. Several types of milled rice are used in this study, namely: Jasmine, Kokuyu, 5-Star, Ifugao, Malagkit, and NFA rice. The captured images are processed using MATLAB R2013a software. There is a USB dongle connected to the router which provided internet connection for online web access. The GizDuino IOT-644 is used for handling the temperature and humidity sensor, and for sending and receiving of data from computer to the cloud storage. The result is compared to the actual moisture content range using a moisture tester for milled rice. Based on results, this study provided accurate data in determining the moisture content of the milled rice.

Tele-operated search robot for human detection using histogram of oriented objects

Disasters such as typhoons, tornadoes, and earthquakes are inevitable. Aftermaths of these disasters include the missing people. Using robots with human detection capabilities to locate the missing people, can dramatically reduce the harm and risk to those who work in such circumstances. This study aims to: design and build a tele-operated robot; implement in MATLAB an algorithm for the detection of humans; and create a database of human identification based on various positions, angles, light intensity, as well as distances from which humans will be identified. Different light intensities were made by using Photoshop to simulate smoke, dust and water drops conditions. After processing the image, the system can indicate either a human is detected or not detected. Testing with bodies covered was also conducted to test the algorithm’s robustness. Based on the results, the algorithm can detect humans with full body shown. For upright and lying positions, detection can happen from 8 feet to 20 feet. For sitting position, detection can happen from 2 feet to 20 feet with slight variances in results because of different lighting conditions. The distances greater than 20 feet, no humans can be processed or false negatives can occur. For bodies covered, the algorithm can detect humans in cases made under given circumstances. On three positions, humans can be detected from 0 degrees to 180 degrees under normal, with smoke, with dust, and with water droplet conditions. This study was able to design and build a tele-operated robot with MATLAB algorithm that can detect humans with an overall precision of 88.30%, from which a database was created for human identification based on various conditions, where humans will be identified.

ISFET with built-in calibration registers through segmented eight-bit binary search in three-point algorithm using FPGA

The calibration of chemical sensors is done manually or wherever the sensor is actually deployed. There are certain procedures that should be accurately followed, otherwise the pH measurement will produce an incorrect output. The main purpose of this paper is to create the circuit of an Ion-Sensitive Field-Effect Transistor (ISFET) with built-in calibration registers through segmented eight-bit binary search in three-point algorithm using FPGA. The circuit used the three-point calibration algorithm and the three standard buffers pH4, pH7, and pH10. The block diagram, schematic diagram, and the number of logic gates were derived after synthesizing the Verilog program in Xilinx/FPGA. An average of 0.30% error was computed to prove reliability of the created circuit in FPGA. Having an ISFET with built-in calibration registers will ease the work of the experts in performing calibrations and could be used as a pH level meter or a remote sensor node in several applications.

Application specific integrated circuit (ASIC) for Ion Sensitive Field Effect Transistor (ISFET) L-Asparagine biosensor

This work presents the design of application specific integrated circuit (ASIC) intended for L-Asparagine measurement, using ion sensitive field effect transistor (ISFET) as biosensor. The L-Asparagine is an amino acid that promotes the growth of cancer cells in leukemia patients. The designed ASIC is comprised of ISFET bridge-type readout circuit, inverting operational amplifier, and peak detector circuit. The L-Asparagine concentrations are translated into pH levels. SPICE simulations showed a sensitivity of −72.78 mV/pH with a linear correlation of almost 1 across the output of the designed ASIC.

Development of a graphics processing unit accelerated stereo vision system for depth estimation

Stereo vision is an important computer vision technique which estimates depth information from two images - similar to how human vision works. However, the computational intensity of stereo vision deters implementation in real-time systems. Recently, massively parallel processors called Graphics Processing Units or GPUs are gaining traction as an acceleration platform to offload compute-intensive tasks to. This paper presents the development of a GPU-accelerated stereo vision system called DepthStream. The DepthStream algorithm uses a combined absolute difference and census cost, cross-based cost aggregation and bitwise fast region voting to achieve error rates up to 3.7× lower than Block-Matching based algorithms in mainstream libraries. Leveraging the parallel processing power that GPUs offer enables the DepthStream algorithm to achieve processing times up to 3.12× faster than Semi-Global Matching based implementations in the same libraries. The study also includes measurements of the accuracy of the estimated distances of the developed stereo vision system. Results show that the estimated depths of the DepthStream stereo vision system are within 4% of the true value for distances less than 6m. Statistical analysis also indicate that there is no significant difference between the stereo vision systems estimated distances and the actual value.

An earthquake activated power interrupting device using a triaxis accelerometer

Philippines is one of the countries situated along the Pacific Ring of Fire that makes it vulnerable to the occurrence of Earthquake and its after effects such as tsunamis. The high probability for an earthquake to hit the country due to its location and due to the unpredictable manifestation of catastrophe motivated the researcher of this study to design an earthquake - activated device that will cut power lines in residential areas especially those which are in nearby bodies of water. The system design of the device was based on the Philippine Earthquake Intensity Scale built in the PHIVOLCS Earthquake simulator including the ranges of ground acceleration and the vibration to the different directions. The proper functionality and reaction of the proposed device was tested and validated in the same simulator.