In this work we propose a workflow to deal with overlaid images—images with superimposed text and company logos—, which is very common in underwater monitoring videos and surveillance camera footage. It is demonstrated that it is possible to use Explaining Artificial Intelligence to improve deep learning models performance for image classification tasks in general. A deep learning model trained to classify metal surface defect, which previously had a low performance, is then evaluated with Layer-wise relevance propagation—an Explaining Artificial Intelligence technique—to identify problems in a dataset that hinder the training of deep learning models in a wide range of applications. Thereafter, it is possible to remove this unwanted information from the dataset—using different approaches: from cutting part of the images to training a Generative Inpainting neural network model—and retrain the model with the new preprocessed images. This proposed methodology improved F1 score in 20% when compared to the original trained dataset, validating the proposed workflow.

In tropical/subtropical regions, the favorable climate associated with the use of agricultural technologies, such as no tillage, minimum cultivation, irrigation, early varieties, desiccants, flowering inducing, and crop rotation, makes agriculture highly dynamic. In this letter, we present the Campo Verde agricultural database. The purpose of creating and sharing these data is to foster advancement of remote sensing technology in areas of tropical agriculture, primarily the development and testing of methods for crop recognition and agricultural mapping. Campo Verde is a municipality of Mato Grosso state, localized in the Cerrado (Brazilian Savanna) biome, in central west Brazil. Soybean, maize, and cotton are the primary crops cultivated in this region. Double cropping systems are widely adopted in this area. There is also livestock and forestry production. Our database provides the land-use classes for 513 fields by month for one Brazilian crop year (between October 2015 and July 2016). This information was gathered during two field campaigns in Campo Verde (December 2015 and May 2016) and by visual interpretation of a time series of Landsat-8/Operational Land Imager (OLI) images using an experienced interpreter. A set of 14 preprocessed synthetic aperture radar Sentinel-1 and 15 Landsat-8/OLI mosaic images is also made available. It is important to promote the use of radar data for tropical agricultural applications, especially because the use of optical remote sensing in these regions is hindered by the high frequency of cloud cover. To demonstrate the utility of our database, results of an experiment conducted using the Sentinel-1 data set are presented.

The presence of weeds in agricultural crops has been one of the problems of greatest interest in recent years as they consume natural resources and negatively affect the agricultural process. For this purpose, a model has been implemented to segment weed in aerial images. The proposed model relies on DeepLabv3 architecture trained upon patches extracted from high-resolution aerial imagery. The dataset employed consisted in 5 high-resolution images that describes a sugar beet agricultural field in Germany. SegNet and U-Net architectures were selected for comparison purposes. Our results demonstrate that balancing of data, together with a greater spatial context leads better results with DeepLabv3 achieving up to 0.89 and 0.81 in terms of AUC and F1-score, respectively.

The monitoring of agricultural activities at a regular basis is crucial to assure that the food production meets the world population demands, which is increasing yearly. Such information can be derived from remote sensing data. In spite of topic’s relevance, not enough efforts have been invested to exploit modern pattern recognition and machine learning methods for agricultural land-cover mapping from multi-temporal, multi-sensor earth observation data. Furthermore, only a small proportion of the works published on this topic relates to tropical/subtropical regions, where crop dynamics is more complicated and difficult to model than in temperate regions. A major hindrance has been the lack of accurate public databases for the comparison of different classification methods. In this context, the aim of the present paper is to share a multi-temporal and multi-sensor benchmark database that can be used by the remote sensing community for agricultural land-cover mapping. Information about crops in situ was collected in Luís Eduardo Magalhães (LEM) municipality, which is an important Brazilian agricultural area, to create field reference data including information about first and second crop harvests. Moreover, a series of remote sensing images was acquired and pre-processed, from both active and passive orbital sensors (Sentinel-1, Sentinel-2/MSI, Landsat-8/OLI), correspondent to the LEM area, along the development of the main annual crops. In this paper, we describe the LEM database (crop field boundaries, land use reference data and pre-processed images) and present the results of an experiment conducted using the Sentinel-1 and Sentinel-2 data.

Remote Sensing (RS) data have been increasingly applied to assess agricultural yield, production and crop condition. In tropical areas, crop dynamics are complex due to multiple agricultural practices such as irrigation, non-tillage, crop rotation and multiple harvest per year. Spatial and temporal information can improve the performance in land-cover and crop type classification tasks. In this context Deep Learning (DL) have emerged as a powerful state-of-the-art technique in the RS community. This work presents a comparative analysis of traditional and DL (supervised and unsupervised) approaches for crop classification on sequences of multitemporal optical and SAR images. Three different approaches are compared: the image stacking approach, which is used as baseline, and two DL based approaches using Autoencoders (AEs) and Convolutional Neural Networks (CNNs). Experiments were carried out in two datasets from two different municipalities in Brazil, Ipuã in São Paulo state and Campo Verde in Mato Grosso state. It is shown that CNN and AE outperformed the traditional approach based on image stacking in terms of Overall Accuracy and Class Accuracy.

Container cranes are of key importance for maritime cargo transportation. The uninterrupted and all-day operation of these container cranes, which directly affects the efficiency of the port, necessitates the continuous inspection of these massive hoisting steel structures. Due to the large size of cranes, the current manual inspections performed by expert climbers are costly, risky, and time-consuming. This motivates further investigations on automated non-destructive approaches for the remote inspection of fatigue-prone parts of cranes. In this paper, we investigate the effectiveness of color space-based and deep learning-based approaches for separating the foreground crane parts from the whole image. Subsequently, three different ML-based algorithms (k-Nearest Neighbors, Random Forest, and Naive Bayes) are employed to detect the rust and repainting areas from detected foreground parts of the crane body. Qualitative and quantitative comparisons of the results of these approaches were conducted. While quantitative evaluation of pixel-based analysis reveals the superiority of the k-Nearest Neighbors algorithm in our experiments, the potential of Random Forest and Naive Bayes for region-based analysis of the defect is highlighted.