Siham Tabik

Evaluating the consistency of the 1982-1999 NDVI trends in the Iberian Peninsula across four time-series derived from the AVHRR sensor: LTDR, GIMMS, FASIR, and PAL-II.

Successive efforts have processed the Advanced Very High Resolution Radiometer (AVHRR) sensor archive to produce Normalized Difference Vegetation Index (NDVI) datasets (i.e., PAL, FASIR, GIMMS, and LTDR) under different corrections and processing schemes. Since NDVI datasets are used to evaluate carbon gains, differences among them may affect nations’ carbon budgets in meeting international targets (such as the Kyoto Protocol). This study addresses the consistency across AVHRR NDVI datasets in the Iberian Peninsula (Spain and Portugal) by evaluating whether their 1982–1999 NDVI trends show similar spatial patterns. Significant trends were calculated with the seasonal Mann-Kendall trend test and their spatial consistency with partial Mantel tests. Over 23% of the Peninsula (N, E, and central mountain ranges) showed positive and significant NDVI trends across the four datasets and an additional 18% across three datasets. In 20% of Iberia (SW quadrant), the four datasets exhibited an absence of significant trends and an additional 22% across three datasets. Significant NDVI decreases were scarce (croplands in the Guadalquivir and Segura basins, La Mancha plains, and Valencia). Spatial consistency of significant trends across at least three datasets was observed in 83% of the Peninsula, but it decreased to 47% when comparing across the four datasets. FASIR, PAL, and LTDR were the most spatially similar datasets, while GIMMS was the most different. The different performance of each AVHRR dataset to detect significant NDVI trends (e.g., LTDR detected greater significant trends (both positive and negative) and in 32% more pixels than GIMMS) has great implications to evaluate carbon budgets. The lack of spatial consistency across NDVI datasets derived from the same AVHRR sensor archive, makes it advisable to evaluate carbon gains trends using several satellite datasets and, whether possible, independent/additional data sources to contrast.

Deep-learning versus OBIA for scattered shrub detection with Google Earth imagery: Ziziphus lotus as case study

There is a growing demand for accurate high-resolution land cover maps in many fields, e.g., in land-use planning and biodiversity conservation. Developing such maps has been traditionally performed using Object-Based Image Analysis (OBIA) methods, which usually reach good accuracies, but require a high human supervision and the best configuration for one image often cannot be extrapolated to a different image. Recently, deep learning Convolutional Neural Networks (CNNs) have shown outstanding results in object recognition in computer vision and are offering promising results in land cover mapping. This paper analyzes the potential of CNN-based methods for detection of plant species of conservation concern using free high-resolution Google EarthTM images and provides an objective comparison with the state-of-the-art OBIA-methods. We consider as case study the detection of Ziziphus lotus shrubs, which are protected as a priority habitat under the European Union Habitats Directive. Compared to the best performing OBIA-method, the best CNN-detector achieved up to 12% better precision, up to 30% better recall and up to 20% better balance between precision and recall. Besides, the knowledge that CNNs acquired in the first image can be re-utilized in other regions, which makes the detection process very fast. A natural conclusion of this work is that including CNN-models as classifiers, e.g., ResNet-classifier, could further improve OBIA methods. The provided methodology can be systematically reproduced for other species detection using our codes available through (https://github.com/EGuirado/CNN-remotesensing).

A snapshot of image pre-processing for convolutional neural networks:case study of MNIST

In the last five years, deep learning methods and particularly Convolutional Neural Networks (CNNs) have exhibited excellent accuracies in many pattern classification problems. Most of the state-of-the-art models apply data-augmentation techniques at the training stage. This paper provides a brief tutorial on data preprocessing and shows its benefits by using the competitive MNIST handwritten digits classification problem. We show and analyze the impact of different preprocessing techniques on the performance of three CNNs, LeNet, Network3 and DropConnect, together with their ensembles. The analyzed transformations are, centering, elastic deformation, translation, rotation and different combinations of them. Our analysis demonstrates that data-preprocessing techniques, such as the combination of elastic deformation and rotation, together with ensembles have a high potential to further improve the state-of-the-art accuracy in MNIST classification.

Automatic handgun detection alarm in videos using deep learning

Current surveillance and control systems still require human supervision and intervention. This work presents a novel automatic handgun detection system in videos appropriate for both, surveillance and control purposes. We reformulate this detection problem into the problem of minimizing false positives and solve it by i) building the key training data-set guided by the results of a deep Convolutional Neural Networks (CNN) classifier and ii) assessing the best classification model under two approaches, the sliding window approach and region proposal approach. The most promising results are obtained by Faster R-CNN based model trained on our new database. The best detector shows a high potential even in low quality youtube videos and provides satisfactory results as automatic alarm system. Among 30 scenes, it successfully activates the alarm after five successive true positives in a time interval smaller than 0.2s, in 27 scenes. We also define a new metric, Alarm Activation Time per Interval (AATpI), to assess the performance of a detection model as an automatic detection system in videos.

Towards highly accurate coral texture images classification using deep convolutional neural networks and data augmentation

The recognition of coral species based on underwater texture images pose a significant difficulty for machine learning algorithms, due to the three following challenges embedded in the nature of this data: 1) datasets do not include information about the global structure of the coral; 2) several species of coral have very similar characteristics; and 3) defining the spatial borders between classes is difficult as many corals tend to appear together in groups. For this reason, the classification of coral species has always required an aid from a domain expert. The objective of this paper is to develop an accurate classification model for coral texture images. Current datasets contain a large number of imbalanced classes, while the images are subject to inter-class variation. We have analyzed 1) several Convolutional Neural Network (CNN) architectures, 2) data augmentation techniques and 3) transfer learning. We have achieved the state-of-the art accuracies using different variations of ResNet on the two current coral texture datasets, EILAT and RSMAS.

Automatic whale counting in satellite images with deep learning

Despite their interest and threat status, the number of whales in world’s oceans remains highly uncertain. Whales detection is normally carried out from costly sighting surveys, acoustic surveys or through high-resolution orthoimages. Since deep convolutional neural networks (CNNs) achieve great performance in object-recognition in images, here we propose a robust and generalizable CNN-based system for automatically detecting and counting whales from space based on open data and tools. A test of the system on Google Earth images in ten global whale-watching hotspots achieved a performance (F1-measure) of 84% in detecting and 97% in counting 80 whales. Applying this cost-effective method worldwide could facilitate the assessment of whale populations to guide conservation actions. Free and global access to high-resolution imagery for conservation purposes would boost this process.

A tuning approach for iterative multiple 3d stencil pipeline on GPUs: Anisotropic Nonlinear Diffusion algorithm as case study

This paper focuses on challenging applications that can be expressed as an iterative pipeline of multiple 3d stencil stages and explores their optimization space on GPUs. For this study, we selected a representative example from the field of digital signal processing, the Anisotropic Nonlinear Diffusion algorithm. An open issue to these applications is to determine the optimal fission/fusion level of the involved stages and whether that combination benefits from data tiling. This implies exploring a large space of all the possible fission/fusion combinations with and without tiling, thus making the process non-trivial. This study provides insights to reduce the optimization tuning space and programming effort of iterative multiple 3d stencils. Our results demonstrate that all combinations that fuse the bottleneck stencil with high halos update cost (

A First Step to Accelerating Fingerprint Matching Based on Deformable Minutiae Clustering

>25\%