Data scarcity is a major challenge when training deep learning (DL) models. DL demands a large amount of data to achieve exceptional performance. Unfortunately, many applications have small or inadequate data to train DL frameworks. Usually, manual labeling is needed to provide labeled data, which typically involves human annotators with a vast background of knowledge. This annotation process is costly, time-consuming, and error-prone. Usually, every DL framework is fed by a significant amount of labeled data to automatically learn representations. Ultimately, a larger amount of data would generate a better DL model and its performance is also application dependent. This issue is the main barrier for many applications dismissing the use of DL. Having sufficient data is the first step toward any successful and trustworthy DL application. This paper presents a holistic survey on state-of-the-art techniques to deal with training DL models to overcome three challenges including small, imbalanced datasets, and lack of generalization. This survey starts by listing the learning techniques. Next, the types of DL architectures are introduced. After that, state-of-the-art solutions to address the issue of lack of training data are listed, such as Transfer Learning (TL), Self-Supervised Learning (SSL), Generative Adversarial Networks (GANs), Model Architecture (MA), Physics-Informed Neural Network (PINN), and Deep Synthetic Minority Oversampling Technique (DeepSMOTE). Then, these solutions were followed by some related tips about data acquisition needed prior to training purposes, as well as recommendations for ensuring the trustworthiness of the training dataset. The survey ends with a list of applications that suffer from data scarcity, several alternatives are proposed in order to generate more data in each application including Electromagnetic Imaging (EMI), Civil Structural Health Monitoring, Medical imaging, Meteorology, Wireless Communications, Fluid Mechanics, Microelectromechanical system, and Cybersecurity. To the best of the authors’ knowledge, this is the first review that offers a comprehensive overview on strategies to tackle data scarcity in DL.
AZ Khalaf, M kassim Haidir, LK Jasim, Iraqi Journal of Science, 2012
In this paper, an approach for object tracking that is inspired from human oculomotor system is proposed and verified experimentally. The developed approach divided into two phases, fast tracking or saccadic phase and smooth pursuit phase. In the first phase, the field of the view is segmented into four regions that are analogue to retinal periphery in the oculomotor system. When the object of interest is entering these regions, the developed vision system responds by changing the values of the pan and tilt angles to allow the object lies in the fovea area and then the second phase will activate. A fuzzy logic method is implemented in the saccadic phase as an intelligent decision maker to select the values of the pan and tilt angle based
... Show MoreIn this study, chemical oxidation was employed for the synthesis of polypyrrole (PPy) nanofiber. Furthermore, PPy has been subjected to treatment using nanoparticles of neodymium oxide (Nd2O3), which were produced and added in a certain ratio. The inquiry centered on the structural characteristics of the blend of polypyrrole and neodymium oxide after their combination. The investigation utilises X-ray diffraction (XRD), FTIR, and Field Emission Scanning Electron Microscopy (FE-SEM) for PPy, 10%, 30%, and 50% by volume of Nd2O3. According to the electrochemical tests, it has been noted that the nanocomposites exhibit a substantial amount of pseudocapacitive activity.
We conducted a theoretical study on the potential use of amorphous hydrogenated silicon (a-Si:H) as the high-index material in quarter-wave-stack Bragg mirrors for cavity quantum electrodynamics applications. Compared to conventionally employed
Diode laser technology is well established for biomedicine applications which demand high-power pulse-wave. They are extensively utilized from medical imaging and testing to surgical therapies and the latest aesthetic processes. For medical therapeutic practices, diode lasers have become the ideal laser source for this particular purpose. In the last previous years, semiconductor laser technology has evolved to produce high-repetitions rate near-infrared pulsed lasers diodes that are dependable, low-cost, portable, and small-weight, about few grams. In this paper, we review the recent development and demonstration of diode laser devices for biomedical applications recorded in the latest years taking into account the power, wavelength, and p
... Show MoreDespite the antiplaque effect of mouth-rinsing with a combination composed of miswak (Salvadora persica L.) and green tea (Camellia sinensis var. assamica) extracts, no data are available regarding its effect on gingival tissue at the molecular level. This pilot study aimed to assess the effect of oral rinsing with this combination on gingival crevicular fluid (GCF) flow and IL-1β levels. Ten subjects rinsed with either the combination, 0.12% chlorhexidine gluconate (CHX) or distilled water without toothbrushing for 4 days after receiving baseline polishing. GCF IL-1β concentration, influx, resting volume and plaque quantity were measured at baseline and after 4 days for each intervention. No significant differences in GCF flow or
... Show MoreThe aim of this work is to explore the thermal performance of a tracked tubular solar still (TSS) with a parabolic trough concentrator in Baghdad (33.27° N, 44.37° E) in September 2022. The present tubular still is distinguished by its hexagonal glass cover. The effect of integrating the TSS with a heat pipe, the still tilt angle (10°, 15°), and the depth of saline water inside the still partitions on the productivity of freshwater are investigated. The results showed that using heat pipe enhances the freshwater productivity by 25%–40% and the efficiency by 25%. For the still integrated with heat pipe, as the water depth is increased from 5.5 to 6.5 cm the productivity of freshwater is incre
In this review, previous studies on the synthesis and characterization of the metal Complexes with paracetamol by elemental analysis, thermal analysis, (IR, NMR and UV-Vis (spectroscopy and conductivity. In reviewing these studies, the authors found that paracetamol can be coordinated through the pair of electrons on the hydroxyl O-atom, carbonyl O-atom, and N-atom of the amide group. If the paracetamol was a monodentate ligand, it will be coordinated by one of the following atoms O-hydroxyl, O-carbonyl or N-amide. But if the paracetamol was bidentate, it is coordinated by atoms (O-carbonyl and N-amide), (O-hydroxyl and N-amide) or (O-carbonyl and O-hydroxyl). The authors also found that free paracetamol and its complexes have antimicrobial
... Show MoreIn this review, previous studies on the synthesis and characterization of the metal Complexes with paracetamol by elemental analysis, thermal analysis, (IR, NMR and UV-Vis (spectroscopy and conductivity. In reviewing these studies, the authors found that paracetamol can be coordinated through the pair of electrons on the hydroxyl O-atom, carbonyl O-atom, and N-atom of the amide group. If the paracetamol was a monodentate ligand, it will be coordinated by one of the following atoms O-hydroxyl, O-carbonyl or N-amide. But if the paracetamol was bidentate, it is coordinated by atoms (O-carbonyl and N-amide), (O-hydroxyl and N-amide) or (O-carbonyl and O-hydroxyl). The authors also found that free paracetamol and its complexes have antimicrobial
... Show More