Ivan Garcia

Bio: Ivan Garcia is an academic researcher from University of Extremadura. The author has contributed to research in topics: Image segmentation & Photovoltaic system. The author has an hindex of 1, co-authored 1 publications receiving 8 citations.

On the detection of solar panels by image processing techniques

Abstract: This paper proposes a solution based on computer vision to detect solar panels in images. It is based on the definition of a feature vector that characterizes portions of images that can be acquired with a standard camera and with no lighting restrictions. The proposal has been applied to a set of images taken in an operating photovoltaic plant and the results obtained demonstrate its validity and robustness. These results are meant to be used in later stages of a procedure for the optimization of energy efficiency.

SSD vs. YOLO for Detection of Outdoor Urban Advertising Panels under Multiple Variabilities

Abstract: This work compares Single Shot MultiBox Detector (SSD) and You Only Look Once (YOLO) deep neural networks for the outdoor advertisement panel detection problem by handling multiple and combined variabilities in the scenes. Publicity panel detection in images offers important advantages both in the real world as well as in the virtual one. For example, applications like Google Street View can be used for Internet publicity and when detecting these ads panels in images, it could be possible to replace the publicity appearing inside the panels by another from a funding company. In our experiments, both SSD and YOLO detectors have produced acceptable results under variable sizes of panels, illumination conditions, viewing perspectives, partial occlusion of panels, complex background and multiple panels in scenes. Due to the difficulty of finding annotated images for the considered problem, we created our own dataset for conducting the experiments. The major strength of the SSD model was the almost elimination of False Positive (FP) cases, situation that is preferable when the publicity contained inside the panel is analyzed after detecting them. On the other side, YOLO produced better panel localization results detecting a higher number of True Positive (TP) panels with a higher accuracy. Finally, a comparison of the two analyzed object detection models with different types of semantic segmentation networks and using the same evaluation metrics is also included.

Automatic Inspection of Photovoltaic Power Plants Using Aerial Infrared Thermography: A Review

Abstract: In recent years, aerial infrared thermography (aIRT), as a cost-efficient inspection method, has been demonstrated to be a reliable technique for failure detection in photovoltaic (PV) systems. This method aims to quickly perform a comprehensive monitoring of PV power plants, from the commissioning phase through its entire operational lifetime. This paper provides a review of reported methods in the literature for automating different tasks of the aIRT framework for PV system inspection. The related studies were reviewed for digital image processing (DIP), classification and deep learning techniques. Most of these studies were focused on autonomous fault detection and classification of PV plants using visual, IRT and aIRT images with accuracies up to 90%. On the other hand, only a few studies explored the automation of other parts of the procedure of aIRT, such as the optimal path planning, the orthomosaicking of the acquired images and the detection of soiling over the modules. Algorithms for the detection and segmentation of PV modules achieved a maximum F1 score (harmonic mean of precision and recall) of 98.4%. The accuracy, robustness and generalization of the developed algorithms are still the main issues of these studies, especially when dealing with more classes of faults and the inspection of large-scale PV plants. Therefore, the autonomous procedure and classification task must still be explored to enhance the performance and applicability of the aIRT method.

Encoder–Decoder Semantic Segmentation Models for Electroluminescence Images of Thin-Film Photovoltaic Modules

Abstract: We consider a series of image segmentation methods based on the deep neural networks in order to perform semantic segmentation of electroluminescence (EL) images of thin-film modules. We utilize the encoder–decoder deep neural network architecture. The framework is general such that it can easily be extended to other types of images (e.g., thermography) or solar cell technologies (e.g., crystalline silicon modules). The networks are trained and tested on a sample of images from a database with 6000 EL images of copper indium gallium diselenide thin film modules. We selected two types of features to extract, shunts and so called “droplets.” The latter feature is often observed in the set of images. Several models are tested using various combinations of encoder–decoder layers, and a procedure is proposed to select the best model. We show exemplary results with the best selected model. Furthermore, we applied the best model to the full set of 6000 images and demonstrate that the automated segmentation of EL images can reveal many subtle features, which cannot be inferred from studying a small sample of images. We believe these features can contribute to process optimization and quality control.

GeoAI for detection of solar photovoltaic installations in the Netherlands

Abstract: National mapping agencies are responsible for creating and maintaining country wide geospatial datasets that are highly accurate and homogenous. The Netherlands’ Cadastre, Land Registry and Mapping Agency, in short, the Kadaster, has created a database of information related to solar installations, using GeoAI. Deep Learning techniques were employed to detect small and medium-scale solar installations on buildings from very high-resolution aerial images for the whole of the Netherlands. The impact of data pre-processing and post-processing are addressed and evaluated. The process was automatized to deal with enormous data and the method was scaled-up nation-wide with the help of cloud solutions. In order to make this information visible, consistent and usable, we built-upon the existing TernausNet; a convolution neural network (CNN) architecture. Model metrics were evaluated after post-processing. The algorithm when used in combination with automated or custom post-processing improves the results. The precision and recall rates of the model for 3 different regions were evaluated and are on average about 0.93 and 0.92 respectively after implementation of post-processing. Use of custom post-processing improves the results by removing the false positives by at least 50%. The final results were compared with the existing national PV register. Overall, the results are not only useful for policy makers to assist them to take the necessary steps in achieving the energy transition goals but also serves as a register for infrastructure planning.

Remote Solar Panels Identification Based on Patterns Localization

Abstract: Solar panels inspection is an essential task that must be routinely performed by solar plant supervisors in the purpose of maintaining an acceptable energy production. Such task requires a regular assessment of all the solar panels installed in the solar farm. This constraint requires assigning an important financial budget as well as a qualified technical staff. However, integrating an automatic assessment system seems more financially beneficial, especially in terms of mitigating the need for manual panels inspection. The proposed inspection system comprises a Unmanned Aerial Vehicle (UAV) provided with a processing unit (based on embedded system architecture) and HD camera. Actually, before skipping to solar panels state evaluation, each panel must be beforehand localized and extracted from the captured video frames (images). In this paper, we propose a standalone system which allows to localize and crop separately solar panels from video frames. Each panel localization is ensured through a well-defined 4 localization patterns, put on the 4 panel corners. These patterns are then localized by means of a specific scanning process which browses the captured image into two directions (horizontal and vertical) and retains all segments whose the structure is characterized by the ration 1:1:1:1:3:1:1:1:1 (describing pattern structure). Basing on the carried out experiments, our conceived approach turned out less computationally expensive compared to the approaches which are based on edge detectors (e.g. Hough transform), linear regression or infrared imagery.