Diulhio De Oliveira

Mr. Diulhio De Oliveira

orcid

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 July 2018 in Sensors
Verified
Journal Article
Open access
PDF file
DODiulhio De Oliveira
Diulhio De Oliveira
MWMarco Wehrmeister
Marco Wehrmeister
The use of Unmanned Aerial Vehicles (UAV) has been increasing over the last few years in many sorts of applications due mainly to the decreasing cost of this technology. One can see the use of the UAV in several civilian applications such as surveillance and search and rescue. Automatic detection of pedestrians in aerial images is a challenging task. The computing vision system must deal with many sources of variability in the aerial images captured with the UAV, e.g., low-resolution images of pedestrians, images captured at distinct angles due to the degrees of freedom that a UAV can move, the camera platform possibly experiencing some instability while the UAV flies, among others. In this work, we created and evaluated different implementations of Pattern Recognition Systems (PRS) aiming at the automatic detection of pedestrians in aerial images captured with multirotor UAV. The main goal is to assess the feasibility and suitability of distinct PRS implementations running on top of low-cost computing platforms, e.g., single-board computers such as the Raspberry Pi or regular laptops without a GPU. For that, we used four machine learning techniques in the feature extraction and classification steps, namely Haar cascade, LBP cascade, HOG + SVM and Convolutional Neural Networks (CNN). In order to improve the system performance (especially the processing time) and also to decrease the rate of false alarms, we applied the Saliency Map (SM) and Thermal Image Processing (TIP) within the segmentation and detection steps of the PRS. The classification results show the CNN to be the best technique with 99.7% accuracy, followed by HOG + SVM with 92.3%. In situations of partial occlusion, the CNN showed 71.1% sensitivity, which can be considered a good result in comparison with the current state-of-the-art, since part of the original image data is missing. As demonstrated in the experiments, by combining TIP with CNN, the PRS can process more than two frames per second (fps), whereas the PRS that combines TIP with HOG + SVM was able to process 100 fps. It is important to mention that our experiments show that a trade-off analysis must be performed during the design of a pedestrian detection PRS. The faster implementations lead to a decrease in the PRS accuracy. For instance, by using HOG + SVM with TIP, the PRS presented the best performance results, but the obtained accuracy was 35 percentage points lower than the CNN. The obtained results indicate that the best detection technique (i.e., the CNN) requires more computational resources to decrease the PRS computation time. Therefore, this work shows and discusses the pros/cons of each technique and trade-off situations, and hence, one can use such an analysis to improve and tailor the design of a PRS to detect pedestrians in aerial images.

1 May 2016
Verified
Conference Paper
Closed access
DODiulhio De Oliveira
Diulhio De Oliveira
MWMarco Wehrmeister
Marco Wehrmeister
People recognition in aerial imagery is a complex problem since it involves cameras angles, which are subject to six degrees of freedom. Traditional classifiers used in pattern matching algorithms are not robust enough to deal with such issues. Thus it is necessary to use other techniques such as deep learning methods, e.g. Convolutional Neural Networks (CNN), in order to cope with such a complex problem. However, in spite of providing potentially better recognition results, deep learning methods demand greater processing time, hindering its use on on-board embedded systems of small Unmanned Aerial Vehicles (UAV). This work is a step towards real-time (and possibly on-board) people recognition on aerial imagery. We propose a new approach based on CNN for real-time people recognition on aerial images obtained from cameras attached to small UAVs. This approach combines images from a low-cost thermal camera to detect candidate objects and CNN for the classification task. We have compared our approach with two traditional pattern matching approaches: saliency map detection and cascade classifiers. The obtained results show that, by using CNN, higher rates of correct classification within 95% of precision are obtained. These results have been obtained within real-time processing of 1.08 fps in the worst case on a hardware without any GPU processing.

25 January 2023 in Sensors
Verified
Journal Article
Open access
PDF file
DODiulhio De Oliveira
Diulhio De Oliveira
BNBogdan Tomoyuki Nassu
Bogdan Tomoyuki Nassu
MWMarco Wehrmeister
Marco Wehrmeister
In this paper, we introduce a one-class learning approach for detecting modifications in assembled printed circuit boards (PCBs) based on photographs taken without tight control over perspective and illumination conditions. Anomaly detection and segmentation are essential for several applications, where collecting anomalous samples for supervised training is infeasible. Given the uncontrolled environment and the huge number of possible modifications, we address the problem as a case of anomaly detection, proposing an approach that is directed towards the characteristics of that scenario, while being well suited for other similar applications. We propose a loss function that can be used to train a deep convolutional autoencoder based only on images of the unmodified board—which allows overcoming the challenge of producing a representative set of samples containing anomalies for supervised learning. We also propose a function that explores higher-level features for comparing the input image and the reconstruction produced by the autoencoder, allowing the segmentation of structures and components that differ between them. Experiments performed on a dataset built to represent real-world situations (which we made publicly available) show that our approach outperforms other state-of-the-art approaches for anomaly segmentation in the considered scenario, while producing comparable results on a more general object anomaly detection task.

7 October 2022 in arXiv
Verified
Preprint
Open access
PDF file
DODiulhio De Oliveira
Diulhio De Oliveira
BNBogdan Tomoyuki Nassu
Bogdan Tomoyuki Nassu
MWMarco Wehrmeister
Marco Wehrmeister
In this paper, we introduce an approach for detecting modifications in assembled printed circuit boards based on photographs taken without tight control over perspective and illumination conditions. One instance of this problem is the visual inspection of gas pumps PCBs, which can be modified by fraudsters wishing to deceive costumers or evade taxes. Given the uncontrolled environment and the huge number of possible modifications, we address the problem as a case of anomaly detection, proposing an approach that is directed towards the characteristics of that scenario, while being well-suited for other similar applications. The proposed approach employs a deep convolutional autoencoder trained to reconstruct images of an unmodified board, but which remains unable to do the same for images showing modifications. By comparing the input image with its reconstruction, it is possible to segment anomalies and modifications in a pixel-wise manner. Experiments performed on a dataset built to represent real-world situations (and which we will make publicly available) show that our approach outperforms other state-of-the-art approaches for anomaly segmentation in the considered scenario, while producing comparable results on the popular MVTec-AD dataset for a more general object anomaly detection task.