A Two-Branch CNN Architecture for Land Cover Classification of PAN and MS Imagery

The use of Very High Spatial Resolution (VHSR) imagery in remote sensing applications is nowadays a current practice whenever fine-scale monitoring of the earth’s surface is concerned. VHSR Land Cover classification, in particular, is currently a well-established tool to support decisions in several domains, including urban monitoring, agriculture, biodiversity, and environmental assessment. Additionally, land cover classification can be employed to annotate VHSR imagery with the aim of retrieving spatial statistics or areas with similar land cover. Modern VHSR sensors provide data at multiple spatial and spectral resolutions, most commonly as a couple of a higher-resolution single-band panchromatic (PAN) and a coarser multispectral (MS) imagery. In the typical land cover classification workflow, the multi-resolution input is preprocessed to generate a single multispectral image at the highest resolution available by means of a pan-sharpening process. Recently, deep learning approaches have shown the advantages of avoiding data preprocessing by letting machine learning algorithms automatically transform input data to best fit the classification task. Following this rationale, we here propose a new deep learning architecture to jointly use PAN and MS imagery for a direct classification without any prior image sharpening or resampling process. Our method, namely M u l t i R e s o L C C , consists of a two-branch end-to-end network which extracts features from each source at their native resolution and lately combine them to perform land cover classification at the PAN resolution. Experiments are carried out on two real-world scenarios over large areas with contrasted land cover characteristics. The experimental results underline the quality of our method while the characteristics of the proposed scenarios underline the applicability and the generality of our strategy in operational settings.

Download Full-text

A Review on Deep Image Contrast Enhancement

SMART MOVES JOURNAL IJOSCIENCE ◽

10.24113/ijoscience.v6i1.258 ◽

2020 ◽

Vol 6 (1) ◽

pp. 4

Author(s):

Puspad Kumar Sharma ◽

Nitesh Gupta ◽

Anurag Shrivastava

Keyword(s):

Image Processing ◽

Deep Learning ◽

Image Enhancement ◽

Research Work ◽

Atmospheric Condition ◽

Input Image ◽

Learning Approaches ◽

Sensing Applications ◽

Image Contrast Enhancement

In image processing applications, one of the main preprocessing phases is image enhancement that is used to produce high quality image or enhanced image than the original input image. These enhanced images can be used in many applications such as remote sensing applications, geo-satellite images, etc. The quality of an image is affected due to several conditions such as by poor illumination, atmospheric condition, wrong lens aperture setting of the camera, noise, etc [2]. So, such degraded/low exposure images are needed to be enhanced by increasing the brightness as well as its contrast and this can be possible by the method of image enhancement. In this research work different image enhancement techniques are discussed and reviewed with their results. The aim of this study is to determine the application of deep learning approaches that have been used for image enhancement. Deep learning is a machine learning approach which is currently revolutionizing a number of disciplines including image processing and computer vision. This paper will attempt to apply deep learning to image filtering, specifically low-light image enhancement. The review given in this paper is quite efficient for future researchers to overcome problems that helps in designing efficient algorithm which enhances quality of the image.

Download Full-text

Automating Visual Blockage Classification of Culverts with Deep Learning

Applied Sciences ◽

10.3390/app11167561 ◽

2021 ◽

Vol 11 (16) ◽

pp. 7561

Author(s):

Umair Iqbal ◽

Johan Barthelemy ◽

Wanqing Li ◽

Pascal Perez

Keyword(s):

Machine Learning ◽

Deep Learning ◽

False Negative ◽

Hydraulic Modeling ◽

Machine Learning Algorithms ◽

Visual Features ◽

Learning Approaches ◽

Learning Models ◽

Conventional Machine

Blockage of culverts by transported debris materials is reported as the salient contributor in originating urban flash floods. Conventional hydraulic modeling approaches had no success in addressing the problem primarily because of the unavailability of peak floods hydraulic data and the highly non-linear behavior of debris at the culvert. This article explores a new dimension to investigate the issue by proposing the use of intelligent video analytics (IVA) algorithms for extracting blockage related information. The presented research aims to automate the process of manual visual blockage classification of culverts from a maintenance perspective by remotely applying deep learning models. The potential of using existing convolutional neural network (CNN) algorithms (i.e., DarkNet53, DenseNet121, InceptionResNetV2, InceptionV3, MobileNet, ResNet50, VGG16, EfficientNetB3, NASNet) is investigated over a dataset from three different sources (i.e., images of culvert openings and blockage (ICOB), visual hydrology-lab dataset (VHD), synthetic images of culverts (SIC)) to predict the blockage in a given image. Models were evaluated based on their performance on the test dataset (i.e., accuracy, loss, precision, recall, F1 score, Jaccard Index, region of convergence (ROC) curve), floating point operations per second (FLOPs) and response times to process a single test instance. Furthermore, the performance of deep learning models was benchmarked against conventional machine learning algorithms (i.e., SVM, RF, xgboost). In addition, the idea of classifying deep visual features extracted by CNN models (i.e., ResNet50, MobileNet) using conventional machine learning approaches was also implemented in this article. From the results, NASNet was reported most efficient in classifying the blockage images with the 5-fold accuracy of 85%; however, MobileNet was recommended for the hardware implementation because of its improved response time with 5-fold accuracy comparable to NASNet (i.e., 78%). Comparable performance to standard CNN models was achieved for the case where deep visual features were classified using conventional machine learning approaches. False negative (FN) instances, false positive (FP) instances and CNN layers activation suggested that background noise and oversimplified labelling criteria were two contributing factors in the degraded performance of existing CNN algorithms. A framework for partial automation of the visual blockage classification process was proposed, given that none of the existing models was able to achieve high enough accuracy to completely automate the manual process. In addition, a detection-classification pipeline with higher blockage classification accuracy (i.e., 94%) has been proposed as a potential future direction for practical implementation.

Download Full-text

Fingerprint Classification Based on Deep Learning Approaches: Experimental Findings and Comparisons

Symmetry ◽

10.3390/sym13050750 ◽

2021 ◽

Vol 13 (5) ◽

pp. 750

Author(s):

Carmelo Militello ◽

Leonardo Rundo ◽

Salvatore Vitabile ◽

Vincenzo Conti

Keyword(s):

Deep Learning ◽

Large Scale ◽

Learning Approaches ◽

Biometric Recognition ◽

Fingerprint Classification ◽

Model Efficiency ◽

Experimental Findings ◽

Processing Techniques

Biometric classification plays a key role in fingerprint characterization, especially in the identification process. In fact, reducing the number of comparisons in biometric recognition systems is essential when dealing with large-scale databases. The classification of fingerprints aims to achieve this target by splitting fingerprints into different categories. The general approach of fingerprint classification requires pre-processing techniques that are usually computationally expensive. Deep Learning is emerging as the leading field that has been successfully applied to many areas, such as image processing. This work shows the performance of pre-trained Convolutional Neural Networks (CNNs), tested on two fingerprint databases—namely, PolyU and NIST—and comparisons to other results presented in the literature in order to establish the type of classification that allows us to obtain the best performance in terms of precision and model efficiency, among approaches under examination, namely: AlexNet, GoogLeNet, and ResNet. We present the first study that extensively compares the most used CNN architectures by classifying the fingerprints into four, five, and eight classes. From the experimental results, the best performance was obtained in the classification of the PolyU database by all the tested CNN architectures due to the higher quality of its samples. To confirm the reliability of our study and the results obtained, a statistical analysis based on the McNemar test was performed.

Download Full-text

Land Cover Classification of Multispectral Remotely Sensed Data Based on Channel Relative Spatial Pattern

International Journal of Innovative Technology and Exploring Engineering - Special Issue ◽

10.35940/ijitee.b7230.019320 ◽

2020 ◽

Vol 9 (3) ◽

pp. 1415-1421

Keyword(s):

Land Use ◽

Random Forest ◽

Land Cover ◽

Spatial Pattern ◽

Land Cover Classification ◽

Urban Region ◽

Significant Issue ◽

Remotely Sensed Data

Land spread grouping of remotely detected pictures includes characterizing the satellite pictures into various land use/land spread classes, for example, water, urban region, crop land, backwoods and so on. To screen the ecological effects. Highlights like shading and surface assume a prevalent job in land spread grouping. Picking an appropriate shading space is a significant issue for shading picture order. The quality of various shading spaces, for example, RGB, HSV, LUV have been coordinated successfully to make sense of the element vector. In this paper, another Channel Relative Spatial Pattern (CRSP) is proposed for separating the surface highlights. The extricated highlights are prepared and tried with Random Forest (RF) classifier. Examinations were directed on IRS LISS IV datasets and the outcomes were assessed dependent on the disarray grid, characterization exactness and Kappa insights. The proposed surface example is additionally contrasted and the (LBP), (LDP) and (LTrP) surface techniques and the precision appraisal results have demonstrated excep

Download Full-text