M. Rahmat Widyanto

First-order derivative-based super-resolution

A single fast super-resolution method based on first-order derivatives from neighbor pixels is proposed. The basic idea of the proposed method is to exploit a first-order derivatives component of six edge directions around a missing pixel, followed by back projection to reduce noise estimated by the difference between simulated and observed images. Using first-order derivatives as a feature, the proposed method is expected to have low computational complexity, and it can theoretically reduce blur, blocking, and ringing artifacts in edge areas compared to previous methods. Experiments were conducted using 900 natural grayscale images from the USC-SIPI Database. We evaluated the proposed and previous methods using peak signal-to-noise ratio, structural similarity, feature similarity, and computation time. Experimental results indicate that the proposed method clearly outperforms other state-of-the-art algorithms such as fast curvature-based interpolation.

Dental classification for periapical radiograph based on multiple fuzzy attribute

Dental classification for periapical radiograph based on multiple fuzzy attribute is proposed, where each tooth is analyzed based on multiple criteria such as area/perimeter ratio and width/height ratio. A classification method on special type of dental image called periapical radiograph is studied and classification is done without speculative classification (in case of ambiguous object), therefore an accurate and assistive result can be obtained due to its capability to handle ambiguous tooth. Experiment results on 78 periapical dental radiographs from University of Indonesia indicates 82.51% total classification accuracy and 84.29% average classification rate per input radiograph. The proposed classification method is planned to be implemented as a submodule for an under developing dental based personal identification system.

Human emotion recognition based on active appearance model and semi-supervised fuzzy C-means

Human emotion recognition is an emerging research area in the field of social signal processing. Facial expression is an important means to detect human emotion. The problem is some facial expressions represent similar emotions. Thus, the recognition must consider the ambiguity in the way human expresses emotions through face. Existing methods do not take into account the level of expressions ambiguity. In our research, we specify face points and display the degree of fuzzy cluster on eight face emotions, namely anger, contempt, disgust, happy, surprise, sadness, fear, and neutral. The proposed methods are based on Active Appearance Model (AAM) and semi-supervised Fuzzy C-means (FCM). We tested the system on Cohn Kanade+ dataset of facial expression which provided eight classes of human emotion. Our methods gain an average accuracy rate of 80.71% and surpass the existing Fuzzy Inference System.

Mix Emotion Recognition from Facial Expression using SVM-CRF Sequence Classifier

Recently, emotion recognition has gained increasing attention in various applications related to Social Signal Processing (SSP) and human affect. The existing research is mainly focused on six basic emotions (happy, sad, fear, disgust, angry, and surprise). However human expresses many kind of emotions, including mix emotion which has not been explored due to its complexity. We model 12 types of mix emotion recognition from facial expression in a sequence of images using two-stages learning which combines Support Vector Machines (SVM) and Conditional Random Fields (CRF) as sequence classifiers. SVM classifies each image frame and produce emotion label output, subsequently it becomes the input for CRF which yields the mix emotion label of the corresponding observation sequence. We evaluate our proposed model on modified image frames of Cohn Kanade+ dataset, and on our own made mix emotion dataset. We also compare our model with the original CRF model, and our model shows a superior performance result.

Inception learning super-resolution

An efficient network for super-resolution, which we refer to as inception learning super-resolution (ILSR), is proposed. We adopt the inception module from GoogLeNet to exploit multiple features from low-resolution images, yet maintain fast training steps. The proposed ILSR network demonstrates low computation time and fast convergence during the training process. It is divided into three parts: feature extraction, mapping, and reconstruction. In feature extraction, we apply the inception module followed by dimensional reduction. Then, we map features using a simple convolutional layer. Finally, we reconstruct the high-resolution component using the inception module and a 1×1 convolutional layer. Experimental results demonstrate that the proposed network can construct sharp edges and clean textures, and reduce computation time by up to three orders of magnitude compared to state-of-the-art methods.

Gaussian Fuzzy Number for STR-DNA Similarity Calculation Involving Familial and Tribal Relationships

We performed locus similarity calculation by measuring fuzzy intersection between individual locus and reference locus and then performed CODIS STR-DNA similarity calculation. The fuzzy intersection calculation enables a more robust CODIS STR-DNA similarity calculation due to imprecision caused by noise produced by PCR machine. We also proposed shifted convoluted Gaussian fuzzy number (SCGFN) and Gaussian fuzzy number (GFN) to represent each locus value as improvement of triangular fuzzy number (TFN) as used in previous research. Compared to triangular fuzzy number (TFN), GFN is more realistic to represent uncertainty of locus information because the distribution is assumed to be Gaussian. Then, the original Gaussian fuzzy number (GFN) is convoluted with distribution of certain ethnic locus information to produce the new SCGFN which more represents ethnic information compared to original GFN. Experiments were done for the following cases: people with family relationships, people of the same tribe, and certain tribal populations. The statistical test with analysis of variance (ANOVA) shows the difference in similarity between SCGFN, GFN, and TFN with a significant level of 95%. The Tukey method in ANOVA shows that SCGFN yields a higher similarity which means being better than the GFN and TFN methods. The proposed method enables CODIS STR-DNA similarity calculation which is more robust to noise and performed better CODIS similarity calculation involving familial and tribal relationships.

Nonlinear Fuzzy Robust PCA on Shape Modelling of Active Appearance Model for Facial Expression Recognition

Automatic facial expression recognition is one of the potential research area in the field of computer vison. It aims to improve the ability of machine to capture social signals in human. Automatic facial expression recognition is still a challenge. We proposed method using contrast limited adaptive histogram equalization (CLAHE) for pre-processing stage then performed feature extraction using active appearance model (AAM) based on nonlinear fuzzy robust principal component analysis (NFRPCA). The feature extraction results will be classified with support vector machine (SVM). Feature points generated AAM based on NFRPCA more adaptive compared to AAM based PCA. Our proposed methods the average accuracy rate reached 96,87% and 93,94% for six and seven basic emotions respectively.

Shape analysis using generalized procrustes analysis on Active Appearance Model for facial expression recognition

Facial expression recognition is an active research area in the field of signal social processing. The goal is to distinguish human emotion. The problem is similar emotion, variation of emotion, and independent object through face image. The existing research using various method for modeling human facial to entirely describe facial expression through face image. We consider to variation analysis of the face image using Generalized Procrustes Analysis (GPA) method. GPA is implied for modeling variation of facial expression. We fit our GPA model exact the position of facial skeleton using Active Appearance Model (AAM). AAM is needed for extract shape feature of face image. Also, we use Gabor to get texture information of face image. The facial expression recognition method is based on Support Vector Machine (SVM). We tested our model with CK+ and Jaffe dataset on six basic emotion: anger, disgust, fear, happy, sad, and surprise. Our method gained accuracy 93.58% for CK+ dataset and 94.7% for Jaffe dataset.

Weighting for DNA Profiling

Identification of individual STR-based individuals is required for the investigation of Disaster Victim Identification and other applications. The DNA identification of an individual with the DNA of both biological parents, father, and mother, would result in a perfect match value, but what if the biological parents of the individual had died. In this research, we proposed a method of identifying DNA against an individual if one or both of the individual parents were absent, so it was necessary to match the individual DNA profiles with DNA profiles of existing family members. The conclusions from the results of individual DNA matching with DNA of family members were proposed using fuzzy inference system with weighted suggestion according to familial closeness.

Fuzzy logic System for DNA Profile Matching with Embedded Ethnic Inference

This paper proposes a novel technique to do probabilistic inference by using simple Fuzzy logic System (FlS), especially ethnic information in DNA profile matching algorithm. By using the allele marker’s distribution probability density function as the membership function in the FlS, the new technique makes it possible to tell the ethnic similarity between two DNA profiles in a fast and simple way. A data acquired from ethnic groups of Indonesia is used to test the technique and produced promising result, being able to indicate higher ethnic similarity score within an ethnic group and lower similarity between ethnic groups. However, further research is needed to further improve the model and warrant the correctness and accuracy, especially because the data is not very discriminative.