Marine life airborne observation using HOG and SVM classifier

Alsahwa, Bassem; Maussang, Frédéric; Garello, R.; Chevallier, Alexis

doi:10.1109/oceans.2016.7761242

Cited by 7 publications

(4 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Madhogaria et al [13,14] proposed a novel vehicle recognition method that used HOG features and MRF fusion in vehicle detection of aerial digital images. Alsahwa et al [15] used the method of extracting HOG features and putting them into SVM classifier in the Marine biometric recognition system to solve the target recognition task under this background.…”

Section: Research Status Based On Hog Feature Extractionmentioning

confidence: 99%

See 1 more Smart Citation

Research on the Style Classification Method of Clothing Commodity Images

Zhang

Song

2021

Lecture Notes on Data Engineering and Communications Technologies

View full text Add to dashboard Cite

E-commerce has a good development trend in commercial digitization. Purchasing clothing products on e-commerce platforms is also the main purchase channel currently selected by consumers. With the development of the times, the public pays more and more attention to the individual pursuit of clothing styles. The current research on the classification of clothing types is relatively complete, but the classification of clothing styles still deserves further research. Based on product style classification, this paper divides 252 suit pictures into two categories: business style and sports style. First, the random walk algorithm segments the foreground of the clothes in the product image from the cluttered background. Then the HOG algorithm is used to extract features from the obtained image data. Finally, the extracted feature vectors are put into the classification model of the support vector machine for binary classification. The result is that the accuracy of style classification after segmentation is better than that without segmentation. It also takes less time to classify after random walk processing.

show abstract

Section: Research Status Based On Hog Feature Extractionmentioning

confidence: 99%

“…g(x) ≥ 1 is in the feasible solution area, and θ (w) is the original function at this time. Combine the above two situations to get a new objective function (15).…”

Section: Svmmentioning

confidence: 99%

Research on the Style Classification Method of Clothing Commodity Images

Zhang

Song

2021

Lecture Notes on Data Engineering and Communications Technologies

View full text Add to dashboard Cite

show abstract

“…Therefore, the contribution of the similarity constraint loss and center loss is validated. To test the target classification performance of the IICL-CNN model, we compared it with InceptionV3 [14], InceptionV4 [15], SVDNet [25], PCANet [26], defogging algorithm [30], Resnet50 [42], and a traditional method using SVM classifier and the histogram of oriented gradients (SVM + HOG) [43]. The test set contains 335 clear ship images and their simulated fog occluded images at 5 occlusion levels resulting in atotal number of 2010 samples.…”

Section: F Ship Target Classificationmentioning

confidence: 99%

An Improved InceptionV3 Network for Obscured Ship Classification in Remote Sensing Images

Liu

2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Ship target classification plays an important role in tasks such as maritime traffic control, maritime target tracking, and military reconnaissance. The complex ocean environment often causes obscuration of the ship targets, thus resulting in low accuracy of the obscured targets. This paper presents a novel target classification algorithm-Improved InceptionV3 and Center Loss CNN (IICL-CNN)-based on the well-established Inception network to improve the accuracy of obscured targets. This algorithm features a new objective function, which is designed to learn common features of both the clear samples and the obscured samples and, in the meantime, reduce the intra-class distance among the obscured samples. Experiments were performed on an optical remote sensing image dataset which consisted of 48,000 ship images in 9 categories. The proposed method demonstrated superior performance on the obscured ship targets compared to the original InceptionV3 model. On average, the accuracy was 4.23%, 5.98%, and 17.48% higher on the ship targets that were occluded by levels of 30%, 50%, and 70%, respectively. Our experimental results showed that the proposed IICL-CNN could effectively improve the accuracy of the ship targets at various occlusion levels. Index Terms-Convolution neural network, deep learning, feature extraction, fog occlusion, image classification, remote sensing image I. INTRODUCTION HIP target classification is an emerging research area which has a broad range of applications in civil and military fields, such as maritime traffic control, onshore search, maritime target tracking, and military reconnaissance. With the developments in computer vision and machine learning, there are many recent studies in this area. Zhu et al. [1] recently proposed a method for ship identification, which extracted multiple highdimensional local features from ship images and then used Support Vector Machine (SVM) for target classification. In order to identify ship targets more accurately, SVM is combined with other methods to improve effectiveness and robustness [2, 3, 4, 5]. Aiming at the sparseness of targets in optical remote sensing images, Li et al. [6] proposed a method for target detection and classification based on morphological

show abstract

“…In the field of environmental monitoring, CI techniques often achieve state-of-the-art results thanks to their ability to learn by examples the complex relations between geographical and weather information and the phenomenon of interest, as well as their capability to cope with noisy or incomplete data [16], [17], [18], [19], [20]. Applications of CI for environmental monitoring include the prediction of adverse conditions [21], time series forecasting [22], [23], and the prediction of power produced from renewable energy sources [8], [24], [25], [26].…”

Section: Literature Reviewmentioning

confidence: 99%

Towards the Prediction of Renewable Energy Unbalance in Smart Grids

Labati

Genovese

Piuri

et al. 2018

2018 IEEE 4th International Forum on Research and Technology for Society and Industry (RTSI)

View full text Add to dashboard Cite

The production of renewable energy is increasing worldwide. To integrate renewable sources in electrical smart grids able to adapt to changes in power usage in heterogeneous local zones, it is necessary to accurately predict the power production that can be achieved from renewable energy sources. By using such predictions, it is possible to plan the power production from non-renewable energy plants to properly allocate the produced power and compensate possible unbalances. In particular, it is important to predict the unbalance between the power produced and the actual power intake at a local level (zones). In this paper, we propose a novel method for predicting the sign of the unbalance between the power produced by renewable sources and the power intake at the local level, considering zones composed of multiple power plants and with heterogeneous characteristics. The method uses a set of historical features and is based on Computational Intelligence techniques able to learn the relationship between historical data and the power unbalance in heterogeneous geographical regions. As a case study, we evaluated the proposed method using data collected by a player in the energy market over a period of seven months. In this preliminary study, we evaluated different configurations of the proposed method, achieving results considered as satisfactory by a player in the energy market.978-1-5386-6282-3/18/$31.00 ©2018 IEEE © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

show abstract

Marine life airborne observation using HOG and SVM classifier

Cited by 7 publications

References 11 publications

Research on the Style Classification Method of Clothing Commodity Images

Research on the Style Classification Method of Clothing Commodity Images

An Improved InceptionV3 Network for Obscured Ship Classification in Remote Sensing Images

Towards the Prediction of Renewable Energy Unbalance in Smart Grids

Contact Info

Product

Resources

About