Deep embedded hybrid CNN–LSTM network for lane detection on NVIDIA Jetson Xavier NX

Kortli, Yassin; Gabsi, Souhir; Voon, Lew F.C. Lew Yan; Jridi, Maher; Merzougui, Mehrez; Atri, Mohamed

doi:10.1016/j.knosys.2021.107941

Cited by 52 publications

(16 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…B. Convolutional neural network : Convolutional neural network (CNN) is a kind of deep learning (DL) algorithm and an efficient feature extraction approach that has demonstrated good results in various signal processing and computer vision applications such as automatic speech recognization, image analysis, wind speed forecasting, object detection, and classification 37 . This CNN has the benefit of combining the classification operation and feature extraction in a single unit, allowing CNN to improve the raw datas' feature extraction at the training phase.…”

Section: Proposed Methodsmentioning

confidence: 99%

“…C. Long short term memory (LSTM) network : LSTM network is an improved version of recurrent neural network (RNN); it has the ability to conquer long and short time lags for predicting tasks 37 . The gradient vanishing issue is considered as one of the major complications of RNN in which it makes the network hard to learn long‐term dependencies.…”

Section: Proposed Methodsmentioning

confidence: 99%

“…extraction approach that has demonstrated good results in various signal processing and computer vision applications such as automatic speech recognization, image analysis, wind speed forecasting, object detection, and classification. 37 This CNN has the benefit of combining the classification operation and feature extraction in a single unit, allowing CNN to improve the raw datas' feature extraction at the training phase. The basic idea behind the CNN is to extract local features from higher-layer and transferring to lower layers to acquire complex features.…”

Section: B Convolutional Neural Network: Convolutional Neural Network...mentioning

confidence: 99%

“…; it has the ability to conquer long and short time lags for predicting tasks. 37 The gradient vanishing issue is considered as one of the major complications of RNN in which it makes the network hard to learn long-term dependencies. This difficulty is resolved by introducing an LSTM network that makes the network remember the data on large delays, leave insignificant data and exhibit data carefully during each time interval.…”

Section: Long Short Term Memory (Lstm) Network: Lstm Network Is An Im...mentioning

confidence: 99%

See 3 more Smart Citations

An efficient hand gesture recognition based on optimal deep embedded hybrid convolutional neural network‐long short term memory network model

Palanisamy¹,

Thangaswamy²

2022

Concurrency and Computation

View full text Add to dashboard Cite

Hand gestures are the nonverbal communication done by individuals who cannot represent their thoughts in form of words. It is mainly used during human-computer interaction (HCI), deaf and mute people interaction, and other robotic interface applications. Gesture recognition is a field of computer science mainly focused on improving the HCI via touch screens, cameras, and kinetic devices. The state-of-art systems mainly used computer vision-based techniques that utilize both the motion sensor and camera to capture the hand gestures in real-time and interprets them via the usage of the machine learning algorithms. Conventional machine learning algorithms often suffer from the different complexities present in the visible hand gesture images such as skin color, distance, light, hand direction, position, and background. In this article, an adaptive weighted multi-scale resolution (AWMSR) network with a deep embedded hybrid convolutional neural network and long short term memory network (hybrid CNN-LSTM) is proposed for identifying the different hand gesture signs with higher recognition accuracy. The proposed methodology is formulated using three steps: input preprocessing, feature extraction, and classification. To improve the complex visual effects present in the input images, a histogram equalization technique is used which improves the size of the gray level pixel in the image and also their occurrence probability. The multi-block local binary pattern (MB-LBP) algorithm is employed for feature extraction which extracts the crucial features present in the image such as hand shape structure feature, curvature feature, and invariant movements. The AWMSR with the deep embedded hybrid CNN-LSTM network is applied in the two-benchmark datasets namely Jochen Triesch static hand posture and NUS hand posture dataset-II to detect its stability in identifying different hand gestures. The weight function of the deep embedded CNN-LSTM architecture is optimized using the puzzle optimization algorithm. The efficiency of the proposed methodology is verified in terms of different performance evaluation metrics such as accuracy, loss, confusion matrix, Intersection over the union, and execution time. The proposed methodology offers recognition accuracy of 97.86% and 98.32% for both datasets.

show abstract

Section: Proposed Methodsmentioning

confidence: 99%

Section: Proposed Methodsmentioning

confidence: 99%

Section: B Convolutional Neural Network: Convolutional Neural Network...mentioning

confidence: 99%

Section: Long Short Term Memory (Lstm) Network: Lstm Network Is An Im...mentioning

confidence: 99%

See 2 more Smart Citations

An efficient hand gesture recognition based on optimal deep embedded hybrid convolutional neural network‐long short term memory network model

Palanisamy¹,

Thangaswamy²

2022

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Las aplicaciones van desde algoritmos de identificación de plantas como se expone [2], de vegetables [3] y frutas [4], muy incidentes en temas de punta como la agricultura de precisión. Así mismo la amplitud del espectro del uso de Deep learning y específicamente, como se evidencia en los trabajos mencionados, de las redes neuronales convolucionales (CNN) abarca reconocimiento de señales electromiografícas [5], detección de materiales peligrosos [6] e incluso en conducción autónoma, para detección de vehículos [7] y líneas guía [8]. De forma tal que incluso sistemas robóticos emplean CNN como parte de su esquema de visión, para aplicaciones como robots cosechadores [9] o de reordenamiento de objetos en bandas transportadoras a nivel industrial [10].…”

Section: Introductionunclassified

Audio Commands Recognition Through Deep Learning for Control Mobile Residential Assistant Robot

Moreno¹,

Estepa²,

Baquero³

2022

Proceedings of the 20th LACCEI International Multi-Conference for Engineering, Education and Technology: “Education, Research A

View full text Add to dashboard Cite

This article exposes the training of a convolutional neural network to recognize voice commands that allow an assistant robot to use in a residential environment. Given the needs for isolation due to health factors, development of care robots that can support a human presents a necessary research focus. For this case, the performance of the network is evaluated, which presents a 94.33% accuracy in the recognition of voice commands, through a virtual environment of a co n ventio na l residence.

show abstract

A study of mechanism-data hybrid-driven method for multibody system via physics-informed neural network

Song,

Wang,

Peng

et al. 2024

Acta Mech. Sin.

View full text Add to dashboard Cite

Deep embedded hybrid CNN–LSTM network for lane detection on NVIDIA Jetson Xavier NX

Cited by 52 publications

References 50 publications

An efficient hand gesture recognition based on optimal deep embedded hybrid convolutional neural network‐long short term memory network model

An efficient hand gesture recognition based on optimal deep embedded hybrid convolutional neural network‐long short term memory network model

Audio Commands Recognition Through Deep Learning for Control Mobile Residential Assistant Robot

A study of mechanism-data hybrid-driven method for multibody system via physics-informed neural network

Contact Info

Product

Resources

About