Content-Aware SLIC Super-Pixels for Semi-Dark Images (SLIC++).

Super-pixels represent perceptually similar visual feature vectors of the image. Super-pixels are the meaningful group of pixels of the image, bunched together based on the color and proximity of singular pixel. Computation of super-pixels is highly affected in terms of accuracy if the image has high pixel intensities, i.e., a semi-dark image is observed. For computation of super-pixels, a widely used method is SLIC (Simple Linear Iterative Clustering), due to its simplistic approach. The SLIC is considerably faster than other state-of-the-art methods. However, it lacks in functionality to retain the content-aware information of the image due to constrained underlying clustering technique. Moreover, the efficiency of SLIC on semi-dark images is lower than bright images. We extend the functionality of SLIC to several computational distance measures to identify potential substitutes resulting in regular and accurate image segments. We propose a novel SLIC extension, namely, SLIC++ based on hybrid distance measure to retain content-aware information (lacking in SLIC). This makes SLIC++ more efficient than SLIC. The proposed SLIC++ does not only hold efficiency for normal images but also for semi-dark images. The hybrid content-aware distance measure effectively integrates the Euclidean super-pixel calculation features with Geodesic distance calculations to retain the angular movements of the components present in the visual image exclusively targeting semi-dark images. The proposed method is quantitively and qualitatively analyzed using the Berkeley dataset. We not only visually illustrate the benchmarking results, but also report on the associated accuracies against the ground-truth image segments in terms of boundary precision. SLIC++ attains high accuracy and creates content-aware super-pixels even if the images are semi-dark in nature. Our findings show that SLIC++ achieves precision of 39.7%, outperforming the precision of SLIC by a substantial margin of up to 8.1%.

Towards Automatic License Plate Detection.

Automatic License Plate Detection (ALPD) is an integral component of using computer vision approaches in Intelligent Transportation Systems (ITS). An accurate detection of vehicles' license plates in images is a critical step that has a substantial impact on any ALPD system's recognition rate. In this paper, we develop an efficient license plate detecting technique through the intelligent combination of Faster R-CNN along with digital image processing techniques. The proposed algorithm initially detects vehicle(s) in the input image through Faster R-CNN. Later, the located vehicle is analyzed by a robust License Plate Localization Module (LPLM). The LPLM module primarily uses color segmentation and processes the HSV image to detect the license plate in the input image. Moreover, the LPLM module employs morphological filtering and dimension analysis to find the license plate. Detailed trials on challenging PKU datasets demonstrate that the proposed method outperforms few recently developed methods by producing high license plates detection accuracy in much less execution time. The proposed work demonstrates a great feasibility for security and target detection applications.

Bimodal Data Fusion of Simultaneous Measurements of EEG and fNIRS during Lower Limb Movements.

Electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) have temporal and spatial characteristics that may complement each other and, therefore, pose an intriguing approach for brain-computer interaction (BCI). In this work, the relationship between the hemodynamic response and brain oscillation activity was investigated using the concurrent recording of fNIRS and EEG during ankle joint movements. Twenty subjects participated in this experiment. The EEG was recorded using 20 electrodes and hemodynamic responses were recorded using 32 optodes positioned over the motor cortex areas. The event-related desynchronization (ERD) feature was extracted from the EEG signal in the alpha band (8-11) Hz, and the concentration change of the oxy-hemoglobin (oxyHb) was evaluated from the hemodynamics response. During the motor execution of the ankle joint movements, a decrease in the alpha (8-11) Hz amplitude (desynchronization) was found to be correlated with an increase of the oxyHb (r = -0.64061, p < 0.00001) observed on the Cz electrode and the average of the fNIRS channels (ch28, ch25, ch32, ch35) close to the foot area representation. Then, the correlated channels in both modalities were used for ankle joint movement classification. The result demonstrates that the integrated modality based on the correlated channels provides a substantial enhancement in ankle joint classification accuracy of 93.01 ± 5.60% (p < 0.01) compared with single modality. These results highlight the potential of the bimodal fNIR-EEG approach for the development of future BCI for lower limb rehabilitation.

Real-Time Stress Assessment Using Sliding Window Based Convolutional Neural Network.

Mental stress has been identified as a significant cause of several bodily disorders, such as depression, hypertension, neural and cardiovascular abnormalities. Conventional stress assessment methods are highly subjective and tedious and tend to lack accuracy. Machine-learning (ML)-based computer-aided diagnosis systems can be used to assess the mental state with reasonable accuracy, but they require offline processing and feature extraction, rendering them unsuitable for real-time applications. This paper presents a real-time mental stress assessment approach based on convolutional neural networks (CNNs). The CNN-based approach afforded real-time mental stress assessment with an accuracy as high as 96%, the sensitivity of 95%, and specificity of 97%. The proposed approach is compared with state-of-the-art ML techniques in terms of accuracy, time utilisation, and quality of features.

Block Compressive Sensing (BCS) Based Low Complexity, Energy Efficient Visual Sensor Platform with Joint Multi-Phase Decoder (JMD).

Devices in a visual sensor network (VSN) are mostly powered by batteries, and in such a network, energy consumption and bandwidth utilization are the most critical issues that need to be taken into consideration. The most suitable solution to such issues is to compress the captured visual data before transmission takes place. Compressive sensing (CS) has emerged as an efficient sampling mechanism for VSN. CS reduces the total amount of data to be processed such that it recreates the signal by using only fewer sampling values than that of the Nyquist rate. However, there are few open issues related to the reconstruction quality and practical implementation of CS. The current studies of CS are more concentrated on hypothetical characteristics with simulated results, rather than on the understanding the potential issues in the practical implementation of CS and its computational validation. In this paper, a low power, low cost, visual sensor platform is developed using an Arduino Due microcontroller board, XBee transmitter, and uCAM-II camera. Block compressive sensing (BCS) is implemented on the developed platform to validate the characteristics of compressive sensing in a real-world scenario. The reconstruction is performed by using the joint multi-phase decoding (JMD) framework. To the best of our knowledge, no such practical implementation using off the shelf components has yet been conducted for CS.

An adaptive learning rate for RBFNN using time-domain feedback analysis.

Radial basis function neural networks are used in a variety of applications such as pattern recognition, nonlinear identification, control and time series prediction. In this paper, the learning algorithm of radial basis function neural networks is analyzed in a feedback structure. The robustness of the learning algorithm is discussed in the presence of uncertainties that might be due to noisy perturbations at the input or to modeling mismatch. An intelligent adaptation rule is developed for the learning rate of RBFNN which gives faster convergence via an estimate of error energy while giving guarantee to the l 2 stability governed by the upper bounding via small gain theorem. Simulation results are presented to support our theoretical development.