A new feature extraction approach based on non linear source separation

Elmannai, Hela; Naceur, Mohamed Saber; Loghmari, Mohamed Anis; Algarni, Abeer D.

doi:10.11591/ijece.v11i5.pp4082-4094

Cited by 4 publications

(4 citation statements)

References 30 publications

(47 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Identifying factors that contributed to injury severity and predicting pedestrian-vehicle accidents were the most studied parts. Many different aspects of human behavior [8], s well as road and environmental design, have been observed to severely influence the frequency and severity of pedestrianvehicle collisions.…”

Section: Related Workmentioning

confidence: 99%

Comparative analysis of influencing factors on pedestrian road accidents

Hafeez

Sheikh²,

Alshammari³

et al. 2023

Bulletin EEI

View full text Add to dashboard Cite

Road accident data includes detailed information about incidents that occurred, such as where they happened, the severity of the accident, and the number of people on the road at the time. Such information is useful in determining the causes of accidents and developing potential countermeasures. This research aims to determine the factors that contribute to pedestrian fatalities and injuries in traffic accidents. This study examined 150 pedestrian-vehicle accidents that took place between 1990 and 2021 in forty countries. Eleven factors have been identified as the major causes of accidents. The categorical principal component analysis (CATPCA) technique is used to reduce the number of dimensions and identify the elements that contribute to accidents. The eleven variables are classified into three groups: human factors, roadway environment, and vehicle attributes. The study found that car speed, weather, lighting, traffic conditions, area types, accident locations, and road conditions all had a significant impact on pedestrian accidents and fatalities. The findings show that a pedestrian's state (walking, running) and intention significantly increase the risk of serious injuries and death. The analysis of the driver's status suggests that the driver's intentions may also play a role in car accidents.

show abstract

Section: Related Workmentioning

confidence: 99%

Comparative analysis of influencing factors on pedestrian road accidents

Hafeez

Sheikh²,

Alshammari³

et al. 2023

Bulletin EEI

View full text Add to dashboard Cite

show abstract

“…In contrast, a multispectral image is one that contains more than one spectral channel. A simple example of a multispectral image is a colour image which contains three channels, corresponding to the red, green and blue wavelength channels of the electromagnetic spectrum [89]- [105].…”

Section: Pan-sharpening Of Spot-5 Imagesmentioning

confidence: 99%

“…Feature extraction allows for a more-accurate detection of features, as a result of the image's enhancements [15], [16], [39], [89], [143]- [145], [148]. The signature of feature classes could be distinguished from complex surface textures.…”

Section: Analysis Of Feature Signaturesmentioning

confidence: 99%

Integrated NIR-HE based SPOT-5 image enhancement method for features preservation and edge detection

Zulkifle

Hassan

Ahmad

et al. 2021

IJEECS

View full text Add to dashboard Cite

Recently, many researchers have directed their attention to methods of predicting shorelines by the use of multispectral images. Thus, a simple and optimised method using image enhancements is proposed to improve the low contrast of the Satellite pour l'Observation de la Terre-5 (SPOT-5) images in the detection of shorelines. The near-infrared (NIR) channel is important in this study to ensure the contrast of the vegetated area and sea classification, due to the high reflectance of leaves in the near infrared wavelength region. This study used five scenes of interest to show the different results in shoreline detection. The results demonstrated that the proposed method performed in an enhanced manner as compared to current methods when dealing with the low contrast ratio of SPOT-5 images. As a result, by utilising the near-infrared histogram equalization (NIR-HE), the contrast of all datasets was efficiently restored, producing a higher efficiency in edge detection, and achieving higher overall accuracy. The improved filtering method showed significantly better shoreline detection results than the other filter methods. It was concluded that this method would be useful for detecting and monitoring the shoreline edge in Tanjung Piai.

show abstract

“…A linear regression is applied for each soil property to establish the relation between the dependent variable y and one or more independent variable X. MLR is based on a functional relationship between spectral properties and reference data. PLSR is a multivariate method that applies statistically derived spectral properties to express the derivation of soil properties from spectral ones [20,21]. Based on projecting predictors and response variables into PLSR factors and correspondent scores [22], a multivariate calibration is established.…”

Section: Introductionmentioning

confidence: 99%

Evaluating the Potentials of PLSR and SVR Models for Soil Properties Prediction Using Field Imaging, Laboratory VNIR Spectroscopy and Their Combination

Karray¹,

Elmannai²,

Toumi³

et al. 2023

Computer Modeling in Engineering &Amp; Sciences

View full text Add to dashboard Cite

Pedo-spectroscopy has the potential to provide valuable information about soil physical, chemical, and biological properties. Nowadays, we may predict soil properties using VNIR field imaging spectra (IS) such as Prisma satellite data or laboratory spectra (LS). The primary goal of this study is to investigate machine learning models namely Partial Least Squares Regression (PLSR) and Support Vector Regression (SVR) for the prediction of several soil properties, including clay, sand, silt, organic matter, nitrate NO3-, and calcium carbonate CaCO 3 , using five VNIR spectra dataset combinations (% IS, % LS) as follows: C1 (0% IS, 100% LS), C2 (20% IS, 80% LS), C3 (50% IS, 50% LS), C4 (80% IS, 20% LS) and C5 (100% IS, 0% LS). Soil samples were collected at bare soils and at the upper (0-30 cm) layer. The data set has been split into a training dataset 80% of the collected data (n = 248) and a validation dataset 20% of the collected data (n = 61). The proposed PLSR and SVR models were trained then tested for each dataset combination. According to our results, SVR outperforms PLSR for both: C1 (0% IS, 100% LS) and C5 (100% IS, 0% LS). For Soil Organic Matter (SOM) prediction, it achieves (R 2 = 0.79%, RMSE = 1.42%) and (R 2 = 0.76%, RMSE = 1.3%), respectively. The data fusion has improved the soil property prediction. The highest improvement was obtained for the SOM property (R 2 = 0.80%, RMSE = 1.39) when using the SVR model and applying the second Combination C2 (20% of IS and 80% LS).

show abstract

A new feature extraction approach based on non linear source separation

Cited by 4 publications

References 30 publications

Comparative analysis of influencing factors on pedestrian road accidents

Comparative analysis of influencing factors on pedestrian road accidents

Integrated NIR-HE based SPOT-5 image enhancement method for features preservation and edge detection

Evaluating the Potentials of PLSR and SVR Models for Soil Properties Prediction Using Field Imaging, Laboratory VNIR Spectroscopy and Their Combination

Contact Info

Product

Resources

About