Abstract
Capsule endoscopy made it possible to observe the inner lumen of the small bowel, but with the cost of a longer duration to process its resulting videos. Therefore, the scientific community has developed several machine learning strategies to help in detecting abnormalities in these videos. The published algorithms are typically trained and evaluated on small sets of images, ultimately not proving to be efficient when applied to full videos. In this experiment, we explored the problem of abnormality classification within an unbalanced dataset of images extracted from video capsule endoscopies, based on a vector feature extracted from the deepest layer of pre-trained Convolution Neural Networks to evaluate the impact of transfer learning with a small number of samples. The results showed that there is a reliable model on the classification task using small portions of data from video capsule endoscopies.

Abstract
The success achieved by using SAR data in the study of the Earth led to a firm commitment from space agencies to develop more and better space-borne SAR sensors. This involvement of the space agencies makes us believe that it is possible to increase the potential of SAR interferometry (InSAR) to near real-time monitoring. Among this ever-increasing number of sensors, the ESA's Sentinel-1 (C-band) mission stands out and appears to be disruptive. This mission is acquiring vast volumes of data making current analyzing approaches inviable. This amount of data can no longer be analyzed and studied using classic methods raising the need to use and create new techniques. We believe that Machine Learning techniques can be the solution to overcome this issue since they allow to train Deep Learning models to automate human processes for a vast volume of data. In this paper, we use deep learning models to automatically find and locate deformation areas in InSAR interferograms without atmospheric correction. We train three state-of-the-art classification models for detection deformation areas, achieving an AUC of 0.864 for the best model (VGG19 for wrapped interferograms). Additionally, we use the same models as encoders to train U-net models, achieving a Dice score of 0.54 for InceptionV3. It is necessary more data to achieve better results in segmentation.

Abstract
With an increasing interest in the digitization effort of ancient manuscripts, ancient character recognition becomes one of the most important areas in the automated document image analysis. In this regard, we propose a Convolutional Neural Network (CNN)-based classifier to recognize the ancient Sundanese characters obtained from a digital collection of Southeast Asian palm leaf manuscripts. In this work, we utilize two different preprocessing techniques for the dataset. The first technique involves the use of geometric transformations, noise background addition, and brightness adjustment to augment the imbalanced samples to be fed into the classifier. The second technique makes use of the Otsu's threshold method to binarize the characters and only uses the usual geometric transformations for the data augmentation. The proposed network with different data augmentation processes is trained on the training set and tested on the testing set. Image binarization from the second technique can outperform the performance of the CNN-based classifier upon the first technique by achieving a testing accuracy of 97.74%. (C) 2021 The Authors. Published by Elsevier B.V.

Abstract
The technological revolution that we have been witnessing recently has allowed components miniaturization and made electronic components accessible. Hyperspectral sensors benefited from these advances and could be mounted on unmanned aerial vehicles, which was unthinkable until recently. This fact significantly increased the applications of hyperspectral data, namely in agriculture, especially in the detection of diseases at an early stage. The vineyard is one of the agricultural sectors that has the most to gain from the use of this type of data, both by the economic value and by the number of diseases the plants are exposed to. The Flavescense dorée is a disease that attacks vineyards and may conduct to a significant loss. Nowadays, the detection of this disease is based on the visual identification of symptoms performed by experts who cover the entire area. However, this work remains tedious and relies only on the human eye, which is a problem since sometimes healthy plants are torn out, while diseased ones are left. If the experts think they have found symptoms, they take samples to send to the laboratory for further analysis. If the test is positive, then the whole vine is uprooted, to limit the spread of the disease. In this context, the use of hyperspectral data will allow the development of new disease detection methods. However, it will be necessary to reduce the volume of data used to make them usable by conventional resources. Fortunately, the advent of machine learning techniques empowered the development of systems that allow better decisions to be made, and consequently save time and money. In this article, a machine learning approach, which is based on an Autoencoder to automatically detect wine disease, is proposed.

Abstract
Botnets are responsible for some of the major malicious traffic on the Internet: DDoS attacks, Mail SPAM, brute force attacks, portscans, and others. Its dangerousness is due to the coordinated amount of infected hosts focusing on a single target. More contributions are in need, considering that (A) ML has been used for cyberattacks identification with better accuracy than standard NIDS equipments, (B) Botnet attacks are one of the most dangerous threats on the Internet. (C) the difficulties in getting representative datasets on some Botnets, and (D) Botnet traffic can be misunderstood by its infrastructure protocol. In this paper, we focus on the identification of Botnet traffic, preventing the communication from the Botmaster to the infected hosts and consequently the Botnet cyberattacks. CICFlowMeter and Machine Learning algorithms were used to analyse Botnet2014 public dataset on four different scenarios: all Botnet traffic on a single class, each class per Botnet traffic and the influence of the IPs address fields Botnet traffic detection. The results shows that Random Forest (RF) and Decision Tree (CART) archived similar accuracies on Botnet traffic classification. Important to say that CART obtained similar results with 10-20% of machine time. The metrics shown that the analysis per specific Botnet has higher accuracy than Any Botnet Traffic analysis. Also, the analysis with the IP addresses and L4 Ports scenario has higher accuracy but lower F1-Score that the equivalent without IP addresses or L4 Ports. At last, Feature Importance results confirms the literature, that Botnet traffic is not a single uniform protocol, but a collection of very different ways of communications between the botmaster and the infected hosts.

Abstract
Wine is the most important product from the Douro Region, in Portugal. Ampelographs are disappearing, and farmers need new solutions to identify grapevine varieties to ensure high-quality standards. The development of methodology capable of automatically identify grapevine are in need. In the scenario, deep learning based methods are emerging as the state-of-art in grapevines classification tasks. In previous work, we verify the deep learning models would benefit from focus classification patches in leaves images areas. Deep learning segmentation methods can be used to find grapevine leaves areas. This paper presents a methodology to segment grapevines images automatically based on the U-net model. A private dataset was used, composed of 733 grapevines images frames extracted from 236 videos collected in a natural environment. The trained model obtained a Dice of 95.6% and an Intersection over Union of 91.6%, results that fully satisfy the need of localise grapevine leaves.