International Journal of Advances in Intelligent Informatics
Not a member yet
235 research outputs found
Sort by
Emotion brain-computer interface using wavelet and recurrent neural networks
Full text linkBrain-Computer Interface (BCI) has an intermediate tool that is usually obtained from EEG signal information. This paper proposed the BCI to control a robot simulator based on three emotions for five seconds by extracting a wavelet function in advance with Recurrent Neural Networks (RNN). Emotion is amongst variables of the brain that can be used to move external devices. BCI's success depends on the ability to recognize one person’s emotions by extracting their EEG signals. One method to appropriately recognize EEG signals as a moving signal is wavelet transformation. Wavelet extracted EEG signal into theta, alpha, and beta wave, and consider them as the input of the RNN technique. Connectivity between sequences is accomplished with Long Short-Term Memory (LSTM). The study also compared frequency extraction methods using Fast Fourier Transform (FFT). The results showed that by extracting EEG signals using Wavelet transformations, we could achieve a confident accuracy of 100% for the training data and 70.54% of new data. While the same RNN configuration without pre-processing provided 39% accuracy, even adding FFT would only increase it to 52%. Furthermore, by using features of the frequency filter, we can increase its accuracy from 70.54% to 79.3%. These results showed the importance of selecting features because of RNNs concern to sequenced its inputs. The use of emotional variables is still relevant for instructions on BCI-based external devices, which provide an average computing time of merely 0.235 seconds
Serial and parallel implementation of Needleman-Wunsch algorithm
Full text linkNeedleman-Wunsch dynamic programming algorithm measures the similarity of the pairwise sequence and finds the optimal pair given the number of sequences. The task becomes nontrivial as the number of sequences to compare or the length of sequences increases. This research aims to parallelize the computation involved in the algorithm to speed up the performance using CUDA. However, there is a data dependency issue due to the property of a dynamic programming algorithm. As a solution, this research introduces the heterogeneous anti-diagonal approach, which benefits from the interaction between the serial implementation on CPU and the parallel implementation on GPU. We then measure and compare the computation time between the proposed approach and a straightforward serial approach that uses CPU only. Measurements of computation times are performed under the same experimental setup and using various pairwise sequences at different lengths. The experiment showed that the proposed approach outperforms the serial method in terms of computation time by approximately three times. Moreover, the computation time of the proposed heterogeneous anti-diagonal approach increases gradually despite the big increments in sequence length, whereas the computation time of the serial approach grows rapidly
An effective hybrid ant lion algorithm to minimize mean tardiness on permutation flow shop scheduling problem
Full text linkThis article aimed to develop an improved Ant Lion algorithm. The objective function was to minimize the mean tardiness on the flow shop scheduling problem with a focus on the permutation flow shop problem (PFSP). The Hybrid Ant Lion Optimization Algorithm (HALO) with local strategy was proposed, and from the total search of the agent, the NEH-EDD algorithm was applied. Moreover, the diversity of the nominee schedule was improved through the use of swap mutation, flip, and slide to determine the best solution in each iteration. Finally, the HALO was compared with some algorithms, while some numerical experiments were used to show the performances of the proposed algorithms. It is important to note that comparative analysis has been previously conducted using the nine variations of the PFSSP problem, and the HALO obtained was compared to other algorithms based on numerical experiments
Improved point center algorithm for K-Means clustering to increase software defect prediction
Full text linkThe k-means is a clustering algorithm that is often and easy to use. This algorithm is susceptible to randomly chosen centroid points so that it cannot produce optimal results. This research aimed to improve the k-means algorithm’s performance by applying a proposed algorithm called point center. The proposed algorithm overcame the random centroid value in k-means and then applied it to predict software defects modules’ errors. The point center algorithm was proposed to determine the initial centroid value for the k-means algorithm optimization. Then, the selection of X and Y variables determined the cluster center members. The ten datasets were used to perform the testing, of which nine datasets were used for predicting software defects. The proposed center point algorithm showed the lowest errors. It also improved the k-means algorithm’s performance by an average of 12.82% cluster errors in the software compared to the centroid value obtained randomly on the simple k-means algorithm. The findings are beneficial and contribute to developing a clustering model to handle data, such as to predict software defect modules more accurately
Classifying Barako coffee leaf diseases using deep convolutional models
Full text linkThis work presents the application of recent Deep Convolutional Models (DCM) to classify Barako leaf diseases. Several selected DCMs performed image classification tasks using Transfer Learning and Fine-Tuning, together with data preprocessing and augmentation. The collected dataset used totals to 4,667. Each labeled into four different classes, which included Coffee Leaf Rust (CLR), Cercospora Leaf Spots (CLS), Sooty Molds (SM), and Healthy Leaves (HL). The DCMs were trained using the partial 4,023 images and validated with the remaining 644. The classification results of the trained models VGG16, Xception, and ResNetV2-152 attained overall accuracies of 97%, 95%, and 91%, respectively. By comparing in terms of True Positive Rate (TPR), we found that Xception has the highest number of correct classifications of CLR, VGG16 with SM, and CLS, while ResNetV2-152 with the lowest TPR for CLR. The evaluated results indicate that the use of Deep Convolutional Models with an adequate amount of data, proper fine-tuning, preprocessing, and transfer learning can yield efficient classifiers for identifying several Barako leaf diseases. This work primarily contributes to the growing field of deep learning, specifically for helping farmers improve their diagnostic process by providing a solution that can automatically classify Barako leaf diseases
Lettuce growth stage identification based on phytomorphological variations using coupled color superpixels and multifold watershed transformation
Full text linkIdentifying the plant's developmental growth stages from seed leaf is crucial to understand plant science and cultivation management deeply. An efficient vision-based system for plant growth monitoring entails optimum segmentation and classification algorithms. This study presents coupled color-based superpixels and multifold watershed transformation in segmenting lettuce plant from complicated background taken from smart farm aquaponic system, and machine learning models used to classify lettuce plant growth as vegetative, head development and for harvest based on phytomorphological profile. Morphological computations were employed by feature extraction of the number of leaves, biomass area and perimeter, convex area, convex hull area and perimeter, major and minor axis lengths of the major axis length the dominant leaf, and length of plant skeleton. Phytomorphological variations of biomass compactness, convexity, solidity, plant skeleton, and perimeter ratio were included as inputs of the classification network. The extracted Lab color space information from the training image set undergoes superpixels overlaying with 1,000 superpixel regions employing K-means clustering on each pixel class. Six-level watershed transformation with distance transformation and minima imposition was employed to segment the lettuce plant from other pixel objects. The accuracy of correctly classifying the vegetative, head development, and harvest growth stages are 88.89%, 86.67%, and 79.63%, respectively. The experiment shows that the test accuracy rates of machine learning models were recorded as 60% for LDA, 85% for ANN, and 88.33% for QSVM. Comparative analysis showed that QSVM bested the performance of optimized LDA and ANN in classifying lettuce growth stages. This research developed a seamless model in segmenting vegetation pixels, and predicting lettuce growth stage is essential for plant computational phenotyping and agricultural practice optimization
Flash communication pattern analysis of fireflies based on computer vision
Full text linkPrevious methods for detecting the flashing behavior of fireflies were using either a photomultiplier tube, a stopwatch, or videography. Limitations and problems are associated with these methods, i.e., errors in data collection and analysis, and it is time-consuming. This study aims to applied a computer vision approach to reduce the time of data collection and analysis as compared to the videography methods by illuminance calculation, time of flash occurrence, and optimize the position coordinate automatically and tracking each firefly individually. The Validation of the approach was performed by comparing the flashing data of male fireflies, Sclerotia aquatilis that was obtained from the analysis of the behavioral video. The pulse duration, flash interval, and flash patterns of S. aquatilis were similar to a reference study. The accuracy ratio of the tracking algorithm for tracking multiple fireflies was 0.94. The time consumption required to analyze the video decreased up to 96.82% and 76.91% when compared with videography and the stopwatch method, respectively. Therefore, this program could be employed as an alternative technique for the study of fireflies flashing behavior
Gabor-enhanced histogram of oriented gradients for human presence detection applied in aerial monitoring
Full text linkIn UAV-based human detection, the extraction and selection of the feature vector are one of the critical tasks to ensure the optimal performance of the detection system. Although UAV cameras capture high-resolution images, human figures' relative size renders persons at very low resolution and contrast. Feature descriptors that can adequately discriminate between local symmetrical patterns in a low-contrast image may improve a human figures' detection in vegetative environments. Such a descriptor is proposed and presented in this paper. Initially, the acquired images are fed to a digital processor in a ground station where the human detection algorithm is performed. Part of the human detection algorithm is the GeHOG feature extraction, where a bank of Gabor filters is used to generate textured images from the original. The local energy for each cell of the Gabor images is calculated to identify the dominant orientations. The bins of conventional HOG are enhanced based on the dominant orientation index and the accumulated local energy in Gabor images. To measure the performance of the proposed features, Gabor-enhanced HOG (GeHOG) and other two recent improvements to HOG, Histogram of Edge Oriented Gradients (HEOG) and Improved HOG (ImHOG), are used for human detection on INRIA dataset and a custom dataset of farmers working in fields captured via unmanned aerial vehicle. The proposed feature descriptor significantly improved human detection and performed better than recent improvements in conventional HOG. Using GeHOG improved the precision of human detection to 98.23% in the INRIA dataset. The proposed feature can significantly improve human detection applied in surveillance systems, especially in vegetative environments
Performances of proposed normalization algorithm for iris recognition
Full text linkIris recognition has very high recognition accuracy in comparison with many other biometric features. The iris pattern is not the same even right and left eye of the same person. It is different and unique. This paper proposes an algorithm to recognize people based on iris images. The algorithm consists of three stages. In the first stage, the segmentation process is using circular Hough transforms to find the region of interest (ROI) of given eye images. After that, a proposed normalization algorithm is to generate the polar images than to enhance the polar images using a modified Daugman’s Rubber sheet model. The last step of the proposed algorithm is to divide the enhance the polar image to be 16 divisions of the iris region. The normalized image is 16 small constant dimensions. The Gray-Level Co-occurrence Matrices (GLCM) technique calculates and extracts the normalized image’s texture feature. Here, the features extracted are contrast, correlation, energy, and homogeneity of the iris. In the last stage, a classification technique, discriminant analysis (DA), is employed for analysis of the proposed normalization algorithm. We have compared the proposed normalization algorithm to the other nine normalization algorithms. The DA technique produces an excellent classification performance with 100% accuracy. We also compare our results with previous results and find out that the proposed iris recognition algorithm is an effective system to detect and recognize person digitally, thus it can be used for security in the building, airports, and other automation in many applications
Fingerprint recognition based on shark smell optimization and genetic algorithm
Full text linkFingerprint recognition is a dominant form of biometric due to its distinctiveness. The study aims to extract and select the best features of fingerprint images, and evaluate the strength of the Shark Smell Optimization (SSO) and Genetic Algorithm (GA) in the search space with a chosen set of metrics. The proposed model consists of seven phases namely, enrollment, image preprocessing by using weighted median filter, feature extraction by using SSO, weight generation by using Chebyshev polynomial first kind (CPFK), feature selection by using GA, creation of a user’s database, and matching features by using Euclidean distance (ED). The effectiveness of the proposed model’s algorithms and performance is evaluated on 150 real fingerprint images that were collected from university students by the ZKTeco scanner at Sulaimani city, Iraq. The system’s performance was measured by three renowned error rate metrics, namely, False Acceptance Rate (FAR), False Rejection Rate (FRR), and Correct Verification Rate (CVR). The experimental outcome showed that the proposed fingerprint recognition model was exceedingly accurate recognition because of a low rate of both FAR and FRR, with a high CVR percentage gained which was 0.00, 0.00666, and 99.334%, respectively. This finding would be useful for improving biometric secure authentication based fingerprint. It is also possibly applied to other research topics such as fraud detection, e-payment, and other real-life applications authentication