Damage Detection Results Research Articles

Output-Only Damage-Detection Method for Concrete Frame Structure Using the Phase Space Reconstruction Method

Abstract Vibration-based techniques for structural health monitoring have garnered increasing attention in recent years, particularly in the context of numerous existing concrete frame structures. However, the detection of minor damages remains a challenging endeavor. Diverging from conventional methods reliant on time, frequency, or time-frequency domains for identification, this study proposes and evaluates an output-only phase space–based algorithms for structural damage detection, which involves reconstructing the dynamic system in phase space. Key to this approach is the introduction of the attractor distance, which is derived from attractor geometry, as a defining feature. Although impact force or ambient vibration typically serve as external excitations, this study employs chaotic excitation to induce vibrations in structures. Numerical and experimental investigations are conducted to assess the efficacy of the proposed phase space reconstruction method in detecting damage. A frame structure modeled as a lumped mass model is simulated and analyzed, exploring various degrees of damage and their effects on identification results. Furthermore, the influence of delay time and Gaussian distributed noise on damage-detection results is scrutinized. Validation experiments on a four-story frame structure subjected to chaotic excitation confirm the utility of the reconstructed attractor for damage detection, particularly in scenarios involving minor damage, with an approximately 5 % reduction in stiffness. The findings suggest that combining trajectory analysis with attractor distance facilitates efficient diagnosis of minor damage. The proposed methodology holds promise for damage identification in more intricate and large-scale concrete frame structures.

Detection and assessment of post-earthquake functional building ceiling damage based on improved YOLOv8

Ceiling is one of the crucial non-structural components in public buildings, but it is susceptible to damage after earthquakes. For important functional buildings, rapid assessment of their safety, economic losses, and recoverability is crucial post-earthquake. However, the current detection and assessment of ceiling damage in earthquake-stricken areas heavily relies on manual efforts. Due to limitations in detection methods and technological levels, these manual efforts struggle to meet the requirements of speed, accuracy, and safety. In response to these challenges, this study proposes a post-earthquake functional building ceiling damage detection and assessment method based on GTS-YOLOv8. Firstly, a dataset containing 1059 real ceiling damage images is established, categorized into two damage types: “Panels Fallen" (PF) and “Complete Destruction" (CD). Next, Ghost convolution is introduced, combined with Swin Transformer blocks, and a novel Slim Neck structure is designed to enhance the YOLOv8 algorithm. The GTS-YOLOv8 object detection model is presented. On the established dataset, GTS-YOLOv8 achieves 92 % mAP50 and 72.2 % mAP50-95, surpassing the baseline YOLOv8 by 2.3 % and 4.1 %, respectively. The study investigates ceiling damage detection under different lighting conditions and shooting distances. The results indicate that the highest detection accuracy is achieved under well-lit conditions and when the shooting perspective covers the entire ceiling area. Finally, based on the research outcomes, a detection and assessment architecture is constructed, and a corresponding system is developed, capable of automatically determining the damage State based on ceiling damage detection results. This architecture provides a foundation for the rapid assessment of building safety, economic losses, and recoverability post-earthquake.

Multi-source dynamic adaptive domain generalization network for crack detection under unknown temperature environment

The dispersion characteristics of Lamb waves are more complex under variable temperature environments, leading to deviation in signal distribution, which significantly reduces the performance of damage detection models. Domain adaptation methods are often applied to damage detection problems with differences in data distribution, but this method requires obtaining the distribution information of the target temperature domain in advance and cannot achieve online damage monitoring under unknown temperature environments. Therefore, this article proposes a multi-source dynamic adaptive generalization network (MDAGN) for crack detection under unknown temperature environments. This network can extend the damage detection model to damage detection tasks under unknown temperatures. The main idea is to optimize the discriminative boundaries of the feature space while mining the generalized damage features and specificity in the temperature domain. Design specific regressors for each source temperature domain to maintain the specificity of the temperature domain, and then fuse the damage detection results of multiple regressors through similarity measurement, enabling the model to extrapolate to temperatures outside the distribution and achieve domain generalization. This article collected structural damage data under unknown temperature environments to verify the effectiveness of this method. The results indicate that this method can effectively monitor cracks under unknown temperature environments, establishing an effective framework for structural damage monitoring under unknown working conditions.

Vibration-based Damage Diagnosis for a Population of Composite Structures Under Uncertainty via Hyper-Sphere and Transfer Component Analysis Based Methods

The problem of vibration-based damage diagnosis, including the detection and type characterization of damages in a population of nominally identical composite structures under operational uncertainty is investigated. This is an important problem for various types of fleets, such as aircraft, drones, fixed wing UAVs and others, which becomes highly challenging if the variability in the composite materials and the manufacturing of the population is significant, while the diagnosis of early-stage damages is pursued. Such a population of 27 composite coupons subjecting to operating variability due to alterations in the temperature, the excitation and the experimental setup is considered in this study. Furthermore, early-stage delamination and impact-induced damages are implemented in 10 of the coupons. Damage diagnosis is herein attempted via two data-driven methods that may account for various types of uncertainty, which are founded on a number of stochastic Multiple-Input Single-Output Transmittance Function AutoRegressive with eXogenous pseudo-eXcitation (MISO-TF-ARX) models. The first employs Hyper Spheres for the representation of the healthy subspace under uncertainty using the parameters of the MISO-TF-ARX models and thus abbreviated as HS-MISO-TF, while the other is based on the Transfer Component Analysis of these parameters, abbreviated as TCA-MISO-TF. The methods’ damage diagnosis performance is experimentally assessed using the above composite coupons and a high number of 7 500 inspection test cases for damage detection, whereas damage characterization is based on 450 test cases. The damage detection and characterization results are presented through Receiver Operating Characteristics curves and confusion matrices, respectively. The obtained results indicate adequate damage detection with the HS-MISO-TF method to achieve 82.1% correct detection for 5% of false alarms, with the TCA-MISO-TF following with 76.4% correct detection. Damage characterization is remarkable with the HS-MISO-TF to correctly characterizing the considered damage type for the 100% of the considered test cases and the TCA-MISO-TF following with 92.2%.

Enhancement of PZT-based damage detection in real-scale post-tensioned anchorage under ambient conditions

Varying ambient temperature significantly affects electromechanical (EM) impedance responses of PZT sensors, leading to inaccurate damage detection results. In this study, we investigate temperature influences on the PZT's EM impedance response and introduce a filtering method to enhance PZT-based damage assessment of a real-scale post-tensioned anchorage. Firstly, the principles of the PZT-based damage identification technique are outlined. Next, the sensitivity of the PZT's EM impedance under simultaneous damage and temperature effects are extensively analyzed and quantified using finite element modelling. Thirdly, a 10-day long-term impedance monitoring of a real-scale anchorage specimen is conducted under ambient temperature conditions and different strand-looseness scenarios. The issues on damage detection performance of the PZT under varying ambient temperature are discussed. Finally, a temperature filtering technique based on principal component analysis is adopted to enhance the accuracy of strand looseness detection in the test anchorage. The numerical and experimental outcomes show that the impact of ambient temperature can overshadow the effect of strand looseness, leading to false damage diagnosis. By removing the influence of temperature on the EM impedance response, the strand looseness in the test anchorage was successfully diagnosed.

Small-sample damage detection of bleacher structure based on GAN and MSS-CNN models

The damage scales and forms of bleacher structure are diverse, and the training by using neural network models may be inadequate when the data sample is limited, resulting in challenges such as overfitting or the inability to generalize new damage scenarios. In order to address the issue of damage detection in bleacher structures with small samples, this paper proposes a multi-scale stride convolutional neural network (MSS-CNN) model. It is trained as a generator and discriminator within a generative adversarial network (GAN) framework. By utilizing GAN to generate data and integrating real data with generated data, the mixed data is input into the MSS-CNN model for training, ultimately yielding damage detection results. In order to validate the effectiveness of this approach, a series of experimental studies are conducted by using a bleacher simulator at Qatar University as the research subject. Furthermore, the model is compared with ResNet, multi-layer perceptron, and support vector machine under identical experimental conditions by comparing real and mixed data. The experimental results consistently demonstrate the superior performance of the MSS-CNN model across multiple experiments. This paper presents a fresh research approach and perspective for addressing the challenge of small-sample damage detection in bleacher structures.

K-Nearest Neighbor Algorithm and Case Base Reasoning on Xenia Car Damage Detection Expert System

PT Astra Daihatsu Motor or commonly abbreviated as ADM is the Sole Agent Brand Holder (ATPM) of Daihatsu cars in Indonesia. Xenia car is one of the most popular cars in Indonesia. Although there are many Xenia car users, it is not uncommon for Xenia cars to experience damage caused by the ignorance of car users, where the car user only knows how to use it but does not know how to maintain the car properly and correctly. Before damage occurs to the car, the car usually experiences several symptoms of damage that the user does not realize. With that, there are often difficulties experienced by users to find out the type of damage to the car. The purpose of applying and designing applications in this study is to apply the Case Based Reasoning method with the K-Nearest Neighbor Algorithm to detect damage to Xenia cars and to design and build applications with the Case Based Reasoning method with the K-Nearest Neighbor Algorithm to detect damage to web-based Xenia cars. This research uses the Research and Development method. Based on the results of research from previous cases, the new case has similarities with 5 cases and the highest similitude value is with the highest type, namely the type of Injector Malfunction damage with a value of 0.625 or around 62.5%. This expert system application can detect and determine the results of Xenia car engine damage detection by applying a method that looks for the closest similarity value of new cases to old cases, namely the Case Based Reasoning method and the K-Nearest Neighbor Algorithm looking for the closest neighbors of the same weight value. This web-based expert system can be used by users to find the results of Xenia Car engine damage detection experienced by determining the symptoms that are available in web-based applications. This web-based application can also provide solutions from the detection results of the type of Xenia car engine damage.

Damage Detection in Glass Fibre Composites Using Cointegrated Hyperspectral Images.

Hyperspectral imaging (HSI) is a remote sensing technique that has been successfully applied for the task of damage detection in glass fibre-reinforced plastic (GFRP) materials. Similarly to other vision-based detection methods, one of the drawbacks of HSI is its susceptibility to the lighting conditions during the imaging, which is a serious issue for gathering hyperspectral data in real-life scenarios. In this study, a data conditioning procedure is proposed for improving the results of damage detection with various classifiers. The developed procedure is based on the concept of signal stationarity and cointegration analysis, and achieves its goal by performing the detection and removal of the non-stationary trends in hyperspectral images caused by imperfect lighting. To evaluate the effectiveness of the proposed method, two damage detection tests have been performed on a damaged GFRP specimen: one using the proposed method, and one using an established damage detection workflow, based on the works of other authors. Application of the proposed procedure in the processing of a hyperspectral image of a damaged GFRP specimen resulted in significantly improved accuracy, sensitivity, and F-score, independently of the type of classifier used.

Time-varying damage detection in beam structures using variational mode decomposition and continuous wavelet transform

Civil engineering structures in operation are likely to suffer damage. During the service life, structural damage evolves gradually from minor to severe. However, most current studies focus on damage localization and quantification of damage severity, and hence little attention has been paid to the detection of time-varying damage. This paper aims to propose a new approach for tracking the damage evolution of beam structures. In this approach, the wavelet threshold method is used at first to denoise the response signal. After that, the variational mode decomposition (VMD) is introduced to decompose the denoised response signal into several mono-components adaptively. A proposed index of wavelet total energy change (WTEC) is then applied to the new decomposed signals to localize damages of beam structures. On a basis of this, a time-varying damage index called wavelet energy change ratio (WECR) is established via the continuous wavelet transform and time window techniques and then applied to the response at the damaged position for tracking damage evolution. The proposed method is verified via a numerical example of a simply supported beam (its length and area of cross section are 5 m and 0.04 m2, respectively) with a sudden and a linear stiffness reduction under 1940 El Centro ground acceleration record. In addition, an experiment on a 10-meters-long steel bridge with abrupt stiffness reduction under a moving load is investigated to demonstrate the effectiveness and accuracy of the proposed time-varying damage detection method. The results show that the index of WTEC is able to localize damages of beam structures effectively whatever they are a single damage position or multiple damage positions. Moreover, the damage evolution can be tracked by the proposed index of WECR accurately even though random noises and end effects pose a great impact on the time-varying damage detection results.

Probabilistic damage identification of FGM structures using model updating procedure based on expansion of incomplete FRF data

The spatial incompleteness and uncertainties of measured data are unavoidable factors that could significantly affect the reliability of damage detection results. It is therefore essential to utilize a method that can effectively deal with the spatial incomplete and contaminated data, and presents probabilistic damage identification results instead of deterministic results. In this regard, a probabilistic damage identification approach for functionally graded materials (FGM) structures using model updating procedure based on expansion of incomplete frequency response function (FRF) data with measurement noise is presented. The model updating procedure is formulated as an optimization scheme which is accomplished by minimizing a cost function based on the changes in expanded FRF data. To expand the incompletely measured FRFs, an iterative order reduction method is performed, which makes the identification resistant to the adverse effects of measurement noise. For the minimization process, we adopt a novel meta-heuristic algorithm called bald eagle search algorithm (BES), which has not yet been tested in the field of model updating. Based on the statistical distributions of the identified stiffness parameters in the damaged and undamaged states, the probability of damage existence (PDE) is established to describe the damage probability for each element. The performance of the proposed model updating procedure is verified using three FGM structures: a simple beam, a two-span beam and a cantilever plate. The statistical results indicate that under a high level of noise (15%), the proposed procedure can provide the prediction of damage localization with a high level of confidence and yield damage estimation results with acceptable errors.

Image-to-Image Translation-Based Structural Damage Data Augmentation for Infrastructure Inspection Using Unmanned Aerial Vehicle

As technology advances, the use of unmanned aerial vehicles (UAVs) and image sensors for structural monitoring and diagnostics is becoming increasingly critical. This approach enables the efficient inspection and assessment of structural conditions. Furthermore, the integration of deep learning techniques has been proven to be highly effective in detecting damage from structural images, as demonstrated in our study. To enable effective learning by deep learning models, a substantial volume of data is crucial, but collecting appropriate instances of structural damage from real-world scenarios poses challenges and demands specialized knowledge, as well as significant time and resources for labeling. In this study, we propose a methodology that utilizes a generative adversarial network (GAN) for image-to-image translation, with the objective of generating synthetic structural damage data to augment the dataset. Initially, a GAN-based image generation model was trained using paired datasets. When provided with a mask image, this model generated an RGB image based on the annotations. The subsequent step generated domain-specific mask images, a critical task that improved the data augmentation process. These mask images were designed based on prior knowledge to suit the specific characteristics and requirements of the structural damage dataset. These generated masks were then used by the GAN model to produce new RGB image data incorporating various types of damage. In the experimental validation conducted across the three datasets to assess the image generation for data augmentation, our results demonstrated that the generated images closely resembled actual images while effectively conveying information about the newly introduced damage. Furthermore, the experimental validation of damage detection with augmented data entailed a comparative analysis between the performance achieved solely with the original dataset and that attained with the incorporation of additional augmented data. The results for damage detection consistently demonstrated that the utilization of augmented data enhanced performance when compared to relying solely on the original images.

Vibration-based building health monitoring using spatio-temporal learning model

Vibration-based building health monitoring is a promising and feasible approach to assess the operational state of building structures in a remote, automated, and continuous fashion; however, efficiently handling high-dimensional vibration signals from multiple sensors and effectively coping with missing/noisy data represent two main technical challenges. In order to overcome these issues, this study proposes a novel, reliable and robust framework, abbreviated CLG-BHM, based on a hybrid deep learning architecture. First, the framework uses a 1D convolutional neural network layer to learn low-dimensional representation vectors of long sensor signals, which preserve underlying structures’ dynamic characteristics. Second, temporal relationships within data are distilled via a Long-Short Term Memory layer. Third, the representation vectors of sensors are aggregated with those of their neighbors in a principled way via a graph attention network layer, resulting in a new latent representation rich in both temporal and spatial information. Finally, the latter is gone through a fully-connected layer to provide damage detection results. The performance and viability of the present method are evidenced via various examples involving a simple lumped mass structure, a semi-rigid steel frame, and an experimental 4-story structure from the literature. Moreover, a robustness study is performed, showing that the method can provide reasonable results with the presence of noisy and missing data.

Reference-free infrared thermography detection with subsurface heating for deep cavity in adhesive of hidden frame glass curtain wall

Since the current infrared thermography (IRT) is not effective in detecting deep and invisible cavities in the silicone structural adhesive of hidden frame glass curtain walls (HFGCW), a reference-free IRT with subsurface heating for the deep cavity is proposed. A near-infrared linear laser with high energy density and high transmission is chosen as the subsurface heating source to directly heat the silicone structural adhesive through the glass. Temporal sequence reconstruction and image enhancement based on reference-free calibration are proposed to reduce thermal inhomogenety and thermal noise and ensure comparable results for damage detection under different environments. The effects of traditional surface heating and subsurface heating are compared and analyzed through numerical simulations. And an evaluated feature, which is the maximal temperature difference feature, derived from temperature difference is used to quantitatively analyze the thermal effect caused by different cavities. The subsurface heating simulation results showed that the highest temperature difference between the region with cavity and defect-free region is up to 88% higher than that of traditional surface heating. The experiments revealed that the deep cavities of different lengths, located at 7 mm, 9 mm, and 11 mm below the glass surface, can be successfully detected using subsurface heating and reference-free calibration. A quadratic linear model is proposed to reflect the relationship between the depths and lengths of cavities and the evaluated feature. In conclusion, the proposed method can protect the HFGCW from deep and invisible cavities which can reduce its adhesion and strength.

AUTOMATIC SURFACE DAMAGE CLASSIFICATION DEVELOPED BASED ON DEEP LEARNING FOR WOODEN ARCHITECTURAL HERITAGE

Abstract. In this paper, we propose a system that automatically classifies the surface damages of wooden architectural cultural heritage based on deep learning algorithms. Commonly, on-site surface damage inspections of cultural heritage are carried out manually by field experts. However, it is difficult to manage cultural heritage because experts are not always onsite to check for damage. To overcome this problem, a deep-learning-based classification method is designed to detect surface damage automatically so that cultural heritage monitoring can be done in real time. The dataset required for the development of the deep learning model utilized 4,000 images taken directly from cultural heritage sites. As a result of a comparative analysis of the performance of four deep learning models for several examples of wooden architectural heritage, the damage detection rate of the deep learning model built in this study showed excellent performance between 94.00 and 96.50%. When gradient-weighted class activation mapping is applied to visualize the damage detection results, the performance of the model with the best performance stood out. The results of this paper are significant as a basic study of the development of a real-time remote damage detection system applicable to cultural heritage sites.

The Wrinkles Characterization in GFRP Composites by Infrared Active Thermography.

An experimental study has been carried out to assess the effectiveness of infrared thermography in wrinkle detection in composite GFRP (Glass Fiber Reinforced Plastic) structures by infrared active thermography. Wrinkles in composite GFRP plates with different weave patterns (twill and satin) have been manufactured with the use of the vacuum bagging method. The different localization of defects in laminates has been taken into account. Transmission and reflection measurement techniques of active thermography have been verified and compared. The section of a turbine blade with a vertical axis of rotation containing post-manufacturing wrinkles has been prepared to verify active thermography measurement techniques in the real structure. In the turbine blade section, the influence of a gelcoat surface on the effectiveness of thermography damage detection has also been taken into account. Straightforward thermal parameters applied in structural health monitoring systems allow an effective damage detection method to be built. The transmission IRT setup allows not only for damage detection and localization in composite structures but also for accurate damage identification. The reflection IRT setup is convenient for damage detection systems coupled with nondestructive testing software. In considered cases, the type of fabric weave has negligible influence on the quality of damage detection results.

Effect of forced assembly on bearing performance of single-lap, countersunk composite bolted joints – Part Ⅰ: Experimental investigation

This paper presented an experimental study on the effect of forced assembly on bearing performance of single-lap, countersunk composite bolted joints. The forced assembly deformation amounts in bearing tests were chosen according to the acoustic emission (AE) damage detection results of the bending test. 3D digital image correlation (DIC) technology was used to measure out-of-plane deformation and surface strain field. Bearing damage was observed through microscopy after bearing tests. It is found that forced assembly causes pre-tension as well as higher loading eccentricity, thus obviously weakening the effective bearing strength of single-lap, countersunk composite joints. Specimens with forced assembly display S-shape out-of-plane bending under effective tensile loading. The damage degree of the non-countersunk hole is close to that of the countersunk hole, which indicates that forced assembly uniformizes bearing load distribution at countersunk hole while aggravates the non-uniformity of bearing load distribution at the non-countersunk hole.

BDD-Net+: A Building Damage Detection Framework Based on Modified Coat-Net

An accurate and fast assessment of damaged buildings following a disaster is critical for planning rescue and reconstruction efforts. The damage assessment by traditional methods is time-consuming and with limited performance. In this paper, we propose an end-to-end deep-learning network named Building Damage Detection Network-plus (BDD-Net+). The BDD-Net+ is based on a combination of convolution layers and transformer blocks. The proposed framework takes advantage of the multiscale residual convolution blocks and self-attention layers. The proposed framework consists of four main steps: (1) data preparation, (2) model training, (3) damage map generation and evaluation, and (4) the use of an explainable artificial intelligence (XAI) framework for understanding and interpretation of the operation model. Experimental results include two representative real-world benchmark datasets (i.e., the Haiti earthquake and the Bata explosion). The obtained results illustrate that BDD-Net+ achieves excellent efficacy in comparison with other state-of-the-art methods. Furthermore, the visualization of the results by XAI shows that BDD-Net+ provides more interpretable and explainable results for damage detection than other studied methods.

Multi-scale entropy based damage detection for thermal protection structures with variation of ambient temperature

The thermal protection structures of supersonic vehicles are vulnerable to damage in the extreme environment of high temperature. With advantages of the large propagation range and high sensitivity to damage, guided waves show great potential for structural health monitoring. However, guided waves are susceptible to ambient temperature change, resulting in low reliability of damage detection results. Therefore, this paper proposed a multi-scale entropy feature extraction method for structural damage detection. Firstly, to eliminate the influence of multi-mode guided waves, a sliding window was used to extract the wave packet of different signals to analyze disturbance caused by damage and temperature. Then, an ant colony optimization algorithm was introduced as a feature fusion method to improve the classification performance. To evaluate the performance of features selected by the ant colony optimization algorithm, K-means clustering algorithm, and silhouette coefficients were utilized to calculate the evaluation function and represent the characteristics of damage. Finally, a set of guided wave signals from 20°C to 40°C were obtained to investigate the influence of temperature variation on the damage feature extraction. The results show that the multi-scale entropy method based on the sliding window can effectively extract damage features in the condition of temperature change.

Physics-guided deep neural network for structural damage identification

The physics-driven method via finite element model and the data-driven methods via supervised learning is commonly used in the analysis of structural damage identification. The finite element model can be susceptible to environmental noise and modeling errors, which leads to model identification results deviating from the real damage. The supervised learning method faces the challenge of insufficient labeled damage data, which limits the learned model in identifying damage scenarios that are not contained in the training data. In this study, a novel physics-guided data-driven method is proposed to incorporate physical perception into the data-driven method for identifying structural damage. Specifically, the physics from the finite element model (e.g., structural dynamic characteristics) is incorporated into the neural network model to guide the damage feature learning from scarce measured data. In addition, a loss function with physical significance is developed to evaluate the discrepancy between the finite element model and the measured data, which guides the neural network to learn the damage features of the real structure. The learned neural network model has the potential to improve the damage detection results. The proposed methodology is validated by a numerical case study of a suspension bridge model and an experimental study of a frame structure.

Damage detection of grotto murals based on lightweight neural network

Non-contact damage detection of grotto murals is a challenging task, due to their minor cracks and subtle defects. In this paper, a lightweight neural network-based grotto mural damage detection algorithm, named Ghost-C3SE YOLOv5, is proposed, which can effectively reduce the network size while obtaining reasonable damage detection results. First, based on the YOLOv5 detection algorithm, the dimension of the convolution layer is reduced, by adjusting the network structure and integrating a Ghost module. Second, the channel attention mechanism is introduced into the feature extraction backbone network, to adjust the weight of each feature according to its importance, which can accelerate the convergence speed of the loss function during the model training process. Finally, the experiment results have shown that the lightweight model Ghost-C3SE YOLOv5, applied on the Pascal Voc dataset, reduces the number of parameters by about 22.55 million while ensuring the detection precision and recall rate. Also, the training time and the model size are reduced by 36.21% and 46.04%, respectively. Precision is increased by 1.29% and recall remains comparative, while the utilization rate of the GPU is improved by 41.55%. This has addressed the shortcomings of laborious process and low accuracy in manual detection of the grotto murals. Furthermore, a real-time detection performance of 44.86 FPS is achieved here for damage detection of murals in the YunGang Grottoes, which has overcome the drawbacks of high model complexity, high computational cost, slow detection speed and low memory utilization of various existing deep learning algorithms.