News
Remote sensing is a type of technology used to measure and monitor biophysical characteristics and human activities. It is an important method for analyzing and mapping the planet's cover, allowing for global, rapid, and reliable data acquisition. Today, it is very common to see remote sensing applied in mapping and monitoring vegetation and agriculture, for example. Thus, new analyses have become possible and relevant within a context of technological innovation combined with forest preservation. Among these are studies that utilize artificial intelligence. Pix Force, in partnership with Eletrobras Furnas, developed a R&D project that uses remote sensing in the monitoring of transmission lines. Check out more details in this text.
Speak with one of our specialists and discover how Pix Force can transform your business
What is remote sensing?
Broadly speaking, remote sensing is a set of techniques and procedures that aim to represent and collect data from the Earth's surface without direct contact with it. Thus, it is possible to gather information about an area, terrain, or object using sensors and instruments in general. This type of technique has become very important nowadays. After all, it is capable of revealing geographic and even historical data about the analyzed spaces, including information such as: • Distribution of forest areas
• Progress of deforestation
• Growth of urban areas
• Among others
Based on this concept, Pix Force's R&D project, in partnership with Eletrobras Furnas, brings a new perspective to the socio-environmental indicators of the analyzed regions. There will be an increased frequency in monitoring land use along the entire length of the transmission lines, but mainly regarding the vegetation and risks surrounding the lines. Quote: Read more: Remote sensing in electric power asset management
The impacts of monitoring
Based on all the data that will be collected, the monitoring predicts positive impacts in all regions and companies involved. A greater incentive is expected for environmental preservation projects in specific areas that are more critical for the electrical system. In addition, the research will use cutting-edge technologies such as artificial neural networks in an unprecedented way for transmission lines in Brazil. The greatest advantage will be in the use of images obtained in different bands of the electromagnetic spectrum, enabling: • Integration of biochemical properties (optical)
• Structural and geometric integration (SAR)
• Altimetric differences in the vegetative surface over time (InSAR)
The processing of SAR images, which aim to assist in terrain variation studies, is a complex process that involves several steps, as explained below.
How does the process work?
The SNAP software presents itself as a free and main image processing tool for the Sentinel 1 system. Processing requires two images, one called master and the other slave, to be digitally aligned and subtracted. With two SAR images obtained, it is possible to get a georeferenced interferogram. However, to achieve this, it is necessary to go through several phases: 1. The first step is to choose the region and polarization through a process called coregistering. The master and slave are precisely aligned for each respective orbital pass. This step is essential for generating the interferogram. Each subswath is formed by several separate strips that must be joined together using the deburst and merge operations. 2. The second step involves filtering various intrusive effects: Earth curvature effect, atmospheric variation effects, and various noise effects. 3. Finally, an operation called reprojection generates the georeferenced output, which can be supplemented to obtain the height variation scale (phase unwrapping). The output product involves the intensity, phase, and correlation bands. The latter serves to evaluate the degree of coherence between pixels, allowing to clearly observe that regions without vegetation have high correlation. Although it is a complex process, obtaining interferograms is considered easy using SNAP tools. To make it more efficient and agile, it will be converted into artificial neural networks, obtaining terrain modeling in transmission lines. Citation: Read more: Meet the 10 best tools related to Remote Sensing
Project Conclusions
This study thus demonstrated the potential of using multitemporal InSAR for land elevation monitoring, as already shown in other works. This, and other terrain modeling methodologies, will be applied throughout the Furnas project. They will help to understand the behavior of the soil and forests along the transmission lines and their complex monitoring structure. Preliminary results indicate that the tools and methodologies are promising, and facilitated in applications with neural networks for large study areas.
Fabio Caraça
Fábio Caraça is the Chief Growth Officer at Pix Force. He leads Pix Force's transformation into a scalable SaaS operation, combining strategic vision, culture, and high-impact execution.
