Transformations in the Photogrammetric Co-Processing of Thermal Infrared Images and RGB Images

Dlesk, A.; Vach, K.; Pavelka, Karel

doi:10.3390/s21155061

Cited by 2 publications

(9 citation statements)

References 20 publications

(33 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A first idea was to transform the information using plane transformation, such as affine or projective transformation. Previously published experiments [28] showed that any plane transformation is not sufficient for close-range purposes. Plane transformation is only sufficient when the images are more-or-less perpendicular to the object, captured over a longer and constant distance from the object, and when the depth of the images is more-or-less constant.…”

Section: Sharpening Methodsmentioning

confidence: 99%

“…TIR images have low resolution compared to the common RGB images [7]. TIR images lack visible features or can even be featureless (if there are no temperature changes in the object scene), which significantly complicates, or makes impossible, SfM processing [28]. Algorithms for the SfM method are not able to identify a sufficient number of key or The structure from motion (SfM) method is a prevalent photogrammetric method [23].…”

Section: Introductionmentioning

confidence: 99%

“…To address the negative factors referred to, several methods of photogrammetric co-processing of TIR and RGB images have been presented in the literature. A review of these methods has been provided in [28]. Generally, for the processing of RGB images, geometric accuracy is used and the results are augmented by corresponding information from TIR images.…”

Section: Introductionmentioning

confidence: 99%

“…The first proposed method is called the "sharpening method" and the second method is called the "reprojection method". The aim of this article is to build on the research published in previous articles [24,28], to describe the sharpening and reprojection methods in detail, and to highlight their specific characteristics. The methods are discussed, and the advantages and disadvantages are pointed out.…”

Section: Introductionmentioning

confidence: 99%

“…The germanium lens steers the radiation to the sensor, which is a grid with resolution much lower than that of sensors of common RGB digital cameras. The resolution of the grid of the TIR cameras is usually from 160 × 120 individual sensors to 640 × 480 individual sensors [28]. Knowing the basic laws of thermography, the software of the TIR cameras recalculates the captured radiation to determine the temperature (possibly in • C).…”

mentioning

confidence: 99%

See 4 more Smart Citations

Photogrammetric Co-Processing of Thermal Infrared Images and RGB Images

Dlesk

Vach²,

Pavelka³

2022

Sensors

Self Cite

View full text Add to dashboard Cite

In some applications of thermography, spatial orientation of the thermal infrared information can be desirable. By the photogrammetric processing of thermal infrared (TIR) images, it is possible to create 2D and 3D results augmented by thermal infrared information. On the augmented 2D and 3D results, it is possible to locate thermal occurrences in the coordinate system and to determine their scale, length, area or volume. However, photogrammetric processing of TIR images is difficult due to negative factors which are caused by the natural character of TIR images. Among the negative factors are the lower resolution of TIR images compared to RGB images and lack of visible features on the TIR images. To eliminate these negative factors, two methods of photogrammetric co-processing of TIR and RGB images were designed. Both methods require a fixed system of TIR and RGB cameras and for each TIR image a corresponding RGB image must be captured. One of the methods was termed sharpening and the result of this method is mainly an augmented orthophoto, and an augmented texture of the 3D model. The second method was termed reprojection and the result of this method is a point cloud augmented by thermal infrared information. The details of the designed methods, as well as the experiments related to the methods, are presented in this article.

show abstract

Section: Sharpening Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

Photogrammetric Co-Processing of Thermal Infrared Images and RGB Images

Dlesk

Vach²,

Pavelka³

2022

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

Flat roof classification and leaks detections by Deep Learning

Zahradník,

Roučka,

Karlovská

2024

CEJ

View full text Add to dashboard Cite

This paper presents an efficient and accurate method for detecting flat roof leaks using a combination of unmanned aerial vehicles (UAVs) and deep learning. The proposed method utilizes a DJI M300 drone equipped with RGB and thermal cameras to capture high-resolution images of the roof. These images are then processed to create orthomosaics and digital elevation models (DEMs). A deep learning model based on the U-NET architecture is then used to segment the roof into different classes, such as PVC foil, windows, and sidewalks. Finally, the damaged insulation is identified by analyzing the temperature distribution within the PVC foil segments. The proposed method has several advantages over traditional inspection methods. It is much faster and more efficient. A UAV can collect images of a large roof in a matter of minutes, while traditional methods can take several days or weeks. The orthomosaics and temperature maps generated by the UAV are much more detailed than the images that can be collected by a human inspector. Third, the UAV-based system is safer. The UAV can collect images of the roof without the need for a human inspector to climb onto the roof, which can be dangerous. The results of this study show that the proposed method is an effective and accurate way to detect flat roof leaks. The deep learning model was able to achieve an overall accuracy of 95% in segmenting the roof into different classes. The method was also able to identify damaged insulation with a high degree of accuracy.

show abstract

Transformations in the Photogrammetric Co-Processing of Thermal Infrared Images and RGB Images

Cited by 2 publications

References 20 publications

Photogrammetric Co-Processing of Thermal Infrared Images and RGB Images

Photogrammetric Co-Processing of Thermal Infrared Images and RGB Images

Flat roof classification and leaks detections by Deep Learning

Contact Info

Product

Resources

About