A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection

Cai, Zhaowei; Fan, Quanfu; Feris, Rogério; Vasconcelos, Nuno

doi:10.1007/978-3-319-46493-0_22

Cited by 1,314 publications

(974 citation statements)

References 41 publications

Supporting

Mentioning

963

Contrasting

Unclassified

Order By: Relevance

“…Recently, it has also been introduced to remote sensing data analysis, such as remote sensing scene classification [43][44][45] and object detection [49][50][51]. However, its superiority has not been explored in SM retrieval.…”

Section: Discussionmentioning

confidence: 99%

Comparing the Performance of Neural Network and Deep Convolutional Neural Network in Estimating Soil Moisture from Satellite Observations

Hang

Liu

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

Soil moisture (SM) plays an important role in hydrological cycle and weather forecasting. Satellite provides the only viable approach to regularly observe large-scale SM dynamics. Conventionally, SM is estimated from satellite observations based on the radiative transfer theory. Recent studies have demonstrated that the neural network (NN) method can retrieve SM with comparable accuracy as conventional methods. Here, we are interested in whether the NN model with more complex structures, namely deep convolutional neural network (DCNN), can bring about further improvement in SM retrievals when compared with the NN model used in recent studies. To achieve this objective, the same input data are used for the DCNN and NN models, including L-band Soil Moisture and Ocean Salinity (SMOS) brightness temperature (TB), C-band Advanced Scatterometer (ASCAT) backscattering coefficients, Moderate Resolution Imaging Spectroradiometer (MODIS) Normalized Difference Vegetation Index (NDVI) and soil temperature. The target SM used to train the DCNN and NN models is the European Center for Medium-range Weather Forecasts Re-Analysis Interim (ERA-Interim) product. The experiment consists of two phases: the learning phase from 1 January to 31 December 2015 and the testing phase from 1 January to 31 December 2016. In the learning phase, we train the DCNN and NN models using the ERA-Interim SM. When evaluation between DCNN and NN against in situ measurements in the testing phase, we find that the temporal correlations between DCNN SM and in situ measurements are higher than those between NN SM and in situ measurements by 6.2% and 2.5% on ascending and descending orbits, respectively. In addition, from the perspective of temporal and spatial dynamics, the simulated SM values by DCNN/NN and the ERA-Interim SM agree relatively well at a global scale. Results suggest that both NN and DCNN models are effective in estimating SM from satellite observations, and DCNN can achieve slightly better performance than NN.

show abstract

Section: Discussionmentioning

confidence: 99%

Comparing the Performance of Neural Network and Deep Convolutional Neural Network in Estimating Soil Moisture from Satellite Observations

Hang

Liu

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Late fusion is used in decision-level, which combines several decision results by voting, finding maximum, or counting mean value [40][41][42][43]. Generally, it takes multiple loss functions to build multiple multi-task optimization problems, and the final output consists of every weighted decision scores.…”

Section: Fusion Strategymentioning

confidence: 99%

Automatic Raft Labeling for Remote Sensing Images via Dual-Scale Homogeneous Convolutional Neural Network

Shi

Zou

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

Raft-culture is a way of utilizing water for farming aquatic product. Automatic raft-culture monitoring by remote sensing technique is an important way to control the crop's growth and implement effective management. This paper presents an automatic pixel-wise raft labeling method based on fully convolutional network (FCN). As rafts are always tiny and neatly arranged in images, traditional FCN method fails to extract the clear boundary and other detailed information. Therefore, a homogeneous convolutional neural network (HCN) is designed, which only consists of convolutions and activations to retain all details. We further design a dual-scale structure (DS-HCN) to integrate higher-level contextual information for accomplishing sea-land segmentation and raft labeling at the same time in a uniform framework. A dataset with Gaofen-1 satellite images was collected to verify the effectiveness of our method. DS-HCN shows a satisfactory performance with a better interpretability and a more accurate labeling result.

show abstract

“…As distinguished from MS-CNN [25], Multi-PerNet is trained once for the classification layer and regression layer. Thus, the loss function is defined as:…”

Section: Figurementioning

confidence: 99%

Vehicle Object Detection in Remote Sensing Imagery Based on Multi-Perspective Convolutional Neural Network

Yang

Zhang

2018

IJGI

View full text Add to dashboard Cite

Most traditional object detection approaches have a deficiency of features, slow detection speed, and high false-alarm rate. To solve these problems, we propose a multi-perspective convolutional neural network (Multi-PerNet) to extract remote sensing imagery features. Regions with CNN features (R-CNN) is a milestone in applying CNN method to object detection. With the help of the great feature extraction and classification performance of CNN, the transformation of object detection problem is realized by the Region Proposal method. Multi-PerNet trains a vehicle object detection model in remote sensing imagery based on Faster R-CNN. During model training, sample images and the labels are inputs, and the output is a detection model. First, Multi-PerNet extracts the feature map. Meanwhile, the area distribution and object-area aspect ratio in the sample images are obtained by k-means clustering. Then, the Faster R-CNN region proposal network generates the candidate windows based on the k-means clustering results. Features of candidate windows can be obtained by mapping candidate windows to the feature map. Finally, the candidate window and its features are inputted to the classifier to be trained to obtain the detection model. Experiment results show that the Multi-PerNet model detection accuracy is improved by 10.1% compared with the model obtained by ZF-net and 1.6% compared with the model obtained by PVANet. Moreover, the model size is reduced by 21.3%.

show abstract

A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection

Cited by 1,314 publications

References 41 publications

Comparing the Performance of Neural Network and Deep Convolutional Neural Network in Estimating Soil Moisture from Satellite Observations

Comparing the Performance of Neural Network and Deep Convolutional Neural Network in Estimating Soil Moisture from Satellite Observations

Automatic Raft Labeling for Remote Sensing Images via Dual-Scale Homogeneous Convolutional Neural Network

Vehicle Object Detection in Remote Sensing Imagery Based on Multi-Perspective Convolutional Neural Network

Contact Info

Product

Resources

About