Narrow Results

Fire detection technology aroused people’s attention increasingly. The main challenge of the fire detection systems is how to reduce false alarms caused by objects like fire’s colors. Most existing algorithms used only features of fire in visual field. In this work, we put forward a new algorithm to detect dynamic fire from the surveillance video based on the combination of radiation domain features model. First, a fire color model is used to extract flame-like pixels as candidate areas in YCbCr space. Second, we convert the candidate regions from the traditional color space into radiation domain in advance by camera calibration. And we use seven features to model the spectral spatio-temporal model of the fire to more accurately characterize the physical and optical properties of the fire. Finally, we choose a two-class SVM classifier to identify the fire from the candidate areas and use a radial basis function kernel to improve the accuracy of the recognition. Two different sets of data are used to validate the algorithm we proposed. And the experimental results indicate that our method performs well in video fire surveillance.

In this paper, a computer vision-based cashew nut grading system has been designed and implemented for classifying different grades of cashew nuts using combined features and machine learning approaches. The important task in the cashew nut grading system is to classify the whole and split down cashew nuts. Since these cashew nuts look very similar from the top view, it is a challenging task to classify the whole cashew nut and split down cashew nuts. Hence, a single-view image of cashew nut has been captured by placing a camera with a distance of 17cm (from the right side of the conveyor belt). The captured red, blue and green images are normalized and converted into hue, saturation and value color space. S channel from HSV image is used for segmentation process using Otsu threshold technique. The total numbers of features extracted are 275 and the features are texture (180), color (90), and shape (5). The constrained optimization-based feature selection method is used and 30 features are selected for further process. The Support Vector Machine (SVM) classifier is used for the classification, and the results obtained from different kernel functions are computed and compared. The 8-layer convolutional neural network (CNN) has been developed in this work for classification and to analyze the performance and accuracy. The accuracy of different machine learning classifiers like SVM 1-1, SVM 1-All and CNN model is also evaluated and compared. The overall accuracy obtained by SVM 1-All with kernel function radial basis for classification is 98.93%.

Paraphrase Recognition systems most often use various lexical, syntactic and semantic features to recognize paraphrases. This paper presents the work done in designing a Support Vector Machine (SVM) based Paraphrase Recognizer and then improving its performance using feature selection strategy. Wrapper method of feature selection has been adopted by combining Genetic Algorithms with Support Vector Machine Classifiers. Experimental results show that applying Feature selection improves the accuracy besides reducing the number of features. The developed paraphrase recognizer has been applied for the Student Answer Evaluation task. The results obtained show that the performance of Answer Evaluation systems which use only half the number of features is comparable to systems using the original feature set.

Depression is a mental disorder that relates to a state of sadness and dejection. It also affects the emotional and physical state of a person. Currently, there are no standard diagnostic tests for depression that are able to produce conclusive results and more over the symptoms of depression are hard to diagnose. A lot of people who are suffering from depression are unaware of their illness. The electroencephalographic (EEG) signals can be used to detect the alterations in the brain’s electrochemical potential. The present work is based on the automated classification of the normal and depression EEG signals. Thus, signal processing methods are used to extract hidden information from the EEG signals. In this work, normal and depression EEG signals are used and discrete wavelet transform (DWT) is performed up to two levels. The features (skewness, energy, kurtosis, standard deviation (SD), mean and entropy) are extracted at the various detailed coefficients levels of the DWT. The extracted features then undergo a statistical analysis method, which is the Student’s t-test that determines the significance of differences in the features. Support Vector Machine classifier with Radial Basis Kernel Function (SVM RBF) was used and the classification accuracy results of 88.9237% was obtained. Hence, this proposed automatic classification system can serve as a useful diagnostic and monitoring tool for detection of depression.

Alcoholism is a complex condition that mainly disturbs the neuronal networks in Central Nervous System (CNS). This disorder not only disturbs the brain, but also affects the behavior, emotions, and cognitive judgements. Electroencephalography (EEG) is a valuable tool to examine the neuropsychiatric disorders like alcoholism. The EEG is a well-established modality to diagnose the electrical activity produced by the populations of neurons in cerebral cortex. However, EEG signals are non-linear in nature; hence very challenging to interpret the valuable information from them using linear methods. Thus, using non-linear methods to analyze EEG signals can be beneficial in order to predict the brain signals condition. This paper presents a computer-aided diagnostic method for the detection of alcoholic EEG signals from normal by employing the non-linear techniques. First, the EEG signals are subjected to six levels of Wavelet Packet Decomposition (WPD) to obtain seven wavebands (delta ($d$), theta ($t$), lower alpha (la), upper alpha (ua), lower beta (lb), upper beta (ub), lower gamma (lg)). From each wavebands (activity bands), 19 non-linear features such as Recurrence Quantification Analysis (RQA) ($R_y^x$), Approximate Entropy ($E_{\mathrm{\text{ap}}}^x$), Energy (${\mathrm{\Omega }}^x$), Fractal Dimension (FD) ($F_D^x$), Permutation Entropy ($E_p^x$), Detrended Fluctuation Analysis (${\alpha }_y^x$), Hurst Exponent ($E_H^x$), Largest Lyapunov Exponent ($E_{\mathrm{\text{LLE}}}^x$), Sample Entropy ($E_s^x$), Shannon’s Entropy ($E_{\mathrm{\text{sh}}}^x$), Renyi’s entropy ($E_r^x$), Tsalli’s entropy ($E_{\mathrm{\text{ts}}}^x$), Fuzzy entropy ($E_f^x$), Wavelet entropy ($E_w^x$), Kolmogorov–Sinai entropy ($E_{\mathrm{\text{ks}}}^x$), Modified Multiscale Entropy ($E_{{\mathrm{\text{mms}}}_y}^x$), Hjorth’s parameters (activity ($S_a^x$), mobility ($H_m^x$), and complexity ($H_c^x$)) are extracted. The extracted features are then ranked using Bhattacharyya, Entropy, Fuzzy entropy-based Max-Relevancy and Min-Redundancy (mRMR), Receiver Operating Characteristic (ROC), $t$-test, and Wilcoxon. These ranked features are given to train Support Vector Machine (SVM) classifier. The SVM classifier with radial basis function (RBF) achieved 95.41% accuracy, 93.33% sensitivity and 97.50% specificity using four non-linear features ranked by Wilcoxon method. In addition, an integrated index called Alcoholic Index (ALCOHOLI) is developed using highly ranked two features for identification of normal and alcoholic EEG signals using a single number. This system is rapid, efficient, and inexpensive and can be employed as an EEG analysis assisting system by clinicians in the detection of alcoholism. In addition, the proposed system can be used in rehabilitation centers to evaluate person with alcoholism over time and observe the outcome of treatment provided for reducing or reversing the impact of the condition on the brain.

Epilepsy is a neurological disorder, and its sudden seizures pose a serious threat to the quality of life of patients. Not only does this condition cause patients to potentially lose control and consciousness during seizures, leading to possible injuries or dangerous situations, but it also has a significant impact on their mental health, triggering issues such as anxiety and depression. An intelligent epilepsy diagnosis process based on electroencephalogram (EEG) signals offers notable advantages. First, it provides noninvasive brainwave signals that accurately monitor and record the patterns and characteristics of epileptic seizures, offering objective diagnostic criteria for doctors. Second, with the use of artificial intelligence and machine learning technologies, large amounts of EEG data can be efficiently analyzed, improving the speed and accuracy of diagnosis and providing timely and effective treatment plans for patients. Additionally, intelligent diagnostic systems can achieve real-time monitoring, promptly alert people to potential epileptic seizures, and provide a safer living environment for patients. In this context, this paper proposes an epilepsy diagnosis method based on a transferred AlexNet model and EEG signals. The main contributions of this paper are as follows. (1) A transfer learning mechanism is incorporated into the AlexNet model through the direct transfer of its neural network structure and the modification of some existing neural network structures, followed by collaborative training with the addition of a domain adaptation layer in the network. This introduced transfer mechanism can address small sample size issues. (2) The traditional AlexNet model suffers from redundant feature extraction, leading to slow training. This paper adds batch normalization (BN) layers after each convolutional layer in the AlexNet model to normalize the features extracted from the convolutional layers. This emphasizes the representation of the important features of EEG signals and enables the lower layers of the network to learn the features needed for EEG signal processing. (3) The transferred AlexNet model proposed in this paper is applied to extract the features of epilepsy EEG signals, and the extracted features are input into a support vector machine (SVM) classifier to obtain epilepsy diagnosis results. Comparative experiments show that the diagnostic method used in this paper yields superior results and shorter training times than those of the competing approaches.

Dialog act (DA) classification is useful to understand the intentions of a human speaker. An effective classification of DA can be exploited for realistic implementation of expert systems. In this work, we investigate DA classification using both acoustic and discourse information for HCRC MapTask data. We extract several different acoustic features and exploit these features using a Hidden Markov Model (HMM) network to classify acoustic information. For discourse feature extraction, we propose a novel parts-of-speech (POS) tagging technique that effectively reduces the dimensionality of discourse features. To classify discourse information, we exploit two classifiers such as a HMM and Support Vector Machine (SVM). We further obtain classifier fusion between HMM and SVM to improve discourse classification. Finally, we perform an efficient decision-level classifier fusion for both acoustic and discourse information to classify 12 different DAs in MapTask data. We obtain 65.2% and 55.4% DA classification rates using acoustic and discourse information, respectively. Furthermore, we obtain combined accuracy of 68.6% for DA classification using both acoustic and discourse information. These accuracy rates of DA classification are either comparable or better than previously reported results for the same data set. For average precision and recall, we obtain accuracy rates of 74.89% and 69.83%, respectively. Therefore, we obtain much better precision and recall rates for most of the classified DAs when compared to existing works on the same HCRC MapTask data set.

Breast cancer is the second most common cancer in females, after lung cancer in the world. In Taiwan, there are about 8500 female suffering from breast cancer every year. The incidence of breast cancer has exceeded cervical cancer and has become the most common female cancer. Immunohistochemistry (IHC) image is widely applied to the diagnosis of breast cancer, but it requires a great deal of manpower and time. The IHC images are scoring as {0+, 1+, 2+ and 3+} corresponding to no staining, weak, moderate and strong staining, respectively. With the growing of image processing techniques, computer-assisted technologies are the best solution to reduce the variability of pathologists evaluation and provide highly specific per-cell information. Therefore, in this paper, we proposed an automatic method to assess the grade of breast cancer in IHC images. The proposed method consists of four steps, including ROI extraction, feature extraction, feature selection and a hierarchical SVM classifier. The hierarchical SVM classifier is utilized to score the IHC images into 0+ (no staining), 1+ (weak), 2+ (moderate) and 3+ (strong staining). According to the experimental results, the proposed method can automatically and effectively asses the score of IHC images; it provides important information to help physicians treat breast cancer.

Colonoscopy has proven to be an active diagnostic tool that examines the lower half of the digestive system’s anomalies. This paper confers a Computer-Aided Detection (CAD) method for polyps from colonoscopy images that helps to diagnose the early stage of Colorectal Cancer (CRC). The proposed method consists primarily of image enhancement, followed by the creation of a saliency map, feature extraction using the Histogram of Oriented-Gradients (HOG) feature extractor, and classification using the Support Vector Machine (SVM). We present an efficient image enhancement algorithm for highlighting clinically significant features in colonoscopy images. The proposed enhancement approach can improve the overall contrast and brightness by minimizing the effects of inconsistent illumination conditions. Detailed experiments have been conducted using the publicly available colonoscopy databases CVC ClinicDB, CVC ColonDB and the ETIS Larib. The performance measures are found to be in terms of precision (91.69%), recall (81.53%), F1-score (86.31%) and F2-score (89.45%) for the CVC ColonDB database and precision (90.29%), recall (61.73%), F1-score (73.32%) and F2-score (82.64%) for the ETIS Larib database. Comparison with the futuristic method shows that the proposed approach surpasses the existing one in terms of precision, F1-score, and F2-score. The proposed enhancement with saliency-based selection significantly reduced the number of search windows, resulting in an efficient polyp detection algorithm.