International Journal of Computational Intelligence and Applications

Non-negative matrix factorization (NMF) is an unsupervised algorithm for clustering where a non-negative data matrix is factorized into (usually) two matrices with the property that all the matrices have no negative elements. This factorization raises the problem of instability, which means whenever we run NMF for the same dataset, we get different factorization. In order to solve the problem of non-uniqueness and to have a more stable solution, we propose a new approach that consists on collaborating different NMF models followed by a consensus. The proposed approach was validated on several datasets and the experimental results showed the effectiveness of our approach which is based on the reducing of standard reconstruction error in NMF model.

Facial expression recognition is an interesting research direction of pattern recognition and computer vision. It has been increasingly used in artificial intelligence, human–computer interaction and security monitoring. In recent years, Convolution Neural Network (CNN) as a deep learning technique and multiple classifier combination method has been applied to gain accurate results in classifying face expressions. In this paper, we propose a multimodal classification approach based on a local texture descriptor representation and a combination of CNN to recognize facial expression. Initially, in order to reduce the influence of redundant information, the preprocessing stage is performed using face detection, face image cropping and texture descriptors of Local Binary Pattern (LBP), Local Gradient Code (LGC), Local Directional Pattern (LDP) and Gradient Direction Pattern (GDP) calculation. Second, we construct a cascade CNN architecture using the multimodal data of each descriptor (CNN_LBP, CNN_LGC, CNN_GDP and CNN_LDP) to extract facial features and classify emotions. Finally, we apply aggregation techniques (sum and product rule) for each modality to combine the four multimodal outputs and thus obtain the final decision of our system. Experimental results using CK$+$ and JAFFE database show that the proposed multimodal classification system achieves superior recognition performance compared to the existing studies with classification accuracy of 97, 93% and 94, 45%, respectively.

Software Product Lines (SPLs) are one of the ways to develop software products by increasing productivity and reducing cost and time in the software development process. In SPLs, each product has many features and it is necessary to consider the optimal and custom features of the products. In fact, selecting key features in SPLs is a challenging process. Feature selection in SPLs is an optimization problem and an NP-Hard problem. One way to select a feature is to use meta-heuristic algorithms modeled on nature, i.e., Bat Algorithm. By modeling bat behavior in prey hunting, a suitable meta-innovative algorithm is considered. This algorithm has important advantages that make it more accurate than conventional methods such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) algorithm. In this paper, to select software product features, idol binary algorithm and artificial neural network are used to identify important features of software products that reduce production costs and time. The experiments in MATLAB software and datasets related to software production lines show that the rate of reduction of target performance error or feature selection cost in software production lines in the proposed method has decreased by 64.17% with increasing population.

The retina is the most important part of the eye. By proper feature extraction, it can be the first step to detect a disease. Morphology of retina blood vessels can be used to identify and classify a disease. A step, such as segmentation and analysis of retinal blood vessels, can assist medical personnel in detecting the severity of a disease. In this paper, vascular segmentation using U-net architecture in the Convolutional Neural Network (CNN) method is proposed to train a sematic segmentation model in retinal blood vessel. In addition, the Contrast Limited Adaptive Histogram Equalization (CLAHE) method is used to increase the contrast of the grayscale and Median Filter is used to obtain better image quality. Data augmentation is also used to maximize the number of datasets owned to make more. The proposed method allows for easier implementation. In this study, the dataset used was STARE with the result of accuracy, sensitivity, specificity, precision, and F1-score that reached 97.64%, 78.18%, 99.20%, 88.77%, and 82.91%.

A large training sample is prerequisite for the successful training of each deep learning model for image classification. Collecting a large dataset is time-consuming and costly, especially for plants. When a large dataset is not available, the challenge is how to use a small or medium size dataset to train a deep model optimally. To overcome this challenge, a novel model is proposed to use the available small size plant dataset efficiently. This model focuses on data augmentation and aims to improve the learning accuracy by oversampling the dataset through representative image patches. To extract the relevant patches, ORB key points are detected in the training images and then image patches are extracted using an innovative algorithm. The extracted ORB image patches are used for dataset augmentation to avoid overfitting during the training phase. The proposed model is implemented using convolutional neural layers, where its structure is based on ResNet architecture. The proposed model is evaluated on a challenging ACHENY dataset. ACHENY is a Chenopodiaceae plant dataset, comprising 27030 images from 30 classes. The experimental results show that the patch-based strategy outperforms the classification accuracy achieved by traditional deep models by 9%.

Sputum smear microscopic examination is an effective, fast, and low-cost technique that is highly specific in areas with a high prevalence of pulmonary tuberculosis. Since manual screening needs trained pathologist in high prevalence zones, the possibility of deploying adequate technicians during the epidemic sessions would be impractical. This condition can cause overburdening and fatigue of working technicians which may tend to reduce the potential efficiency of Tuberculosis (TB) diagnosis. Hence, automation of sputum inspection is the most appropriate aspect in TB outbreak zones to maximize the detection ability. Sputum collection, smear preparing, staining, interpreting smears, and reporting of TB severity are all part of the diagnosis of tuberculosis. This study has analyzed the risk of automating TB severity grading. According to the findings of the analysis, numerous TB-positive cases do not fit into the standard TB severity grade while applying direct rule-driven strategy. The manual investigation, on the other hand, arbitrarily labels the TB grade on those cases. To counter the risk of automation, a fuzzy-based Tuberculosis Severity Level Categorizing Algorithm (TSLCA) is introduced to eliminate uncertainty in classifying the level of TB infection. TSLCA introduces the weight factors, which are dependent on the existence of maximum number of Acid-Fast Bacilli (AFB) per microscopic Field of View (FOV). The fuzzification and defuzzification operations are carried out using the triangular membership function. In addition, the $\alpha$-cut approach is used to eliminate the ambiguity in TB severity grading. Several uncertain TB microscopy screening reports are tested using the proposed TSLCA. Based on the experimental results, it is observed that the TB grading by TSLCA is consistent, error-free, significant and fits exactly into the standard criterion. As a result of the proposed TSLCA, the uncertainty of grading is eliminated and the reliability of tuberculosis diagnosis is ensured when adapting automatic diagnosis.