Mitigating thermal‐induced image drift for videogrammetric technique in support of structural monitoring applications

Zhou, H. F.; Li, Zhao Yi; Lu, Lin Jun; Ni, Yiqing

doi:10.1002/stc.2869

Cited by 3 publications

(5 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared with existing methods for compensating camera-induced thermal errors, the method described in this paper has several advantages [17][18][19][20]. First, it eliminates the need for camera preheating, allowing for measurements to be conducted immediately.…”

Section: Discussionmentioning

confidence: 99%

“…Huafei Zhou used a static target to calibrate the thermal effects at different temperatures, fitting a first-order polynomial model for pixel displacement. He detailed the thermal-induced variability of polynomial parameters, creating a correlation model between these parameters and camera temperature [19]. This method of using different models for different temperatures proves more effective.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Improving Measurement Accuracy of Deep Hole Measurement Instruments through Perspective Transformation

Zhao,

Du,

2024

Sensors

View full text Add to dashboard Cite

Deep hole measurement is a crucial step in both deep hole machining and deep hole maintenance. Single-camera vision presents promising prospects in deep hole measurement due to its simple structure and low-cost advantages. However, the measurement error caused by the heating of the imaging sensor makes it difficult to achieve the ideal measurement accuracy. To compensate for measurement errors induced by imaging sensor heating, this study proposes an error compensation method for laser and vision-based deep hole measurement instruments. This method predicts the pixel displacement of the entire field of view using the pixel displacement of fixed targets within the camera’s field of view and compensates for measurement errors through a perspective transformation. Theoretical analysis indicates that the perspective projection matrix changes due to the heating of the imaging sensor, which causes the thermally induced measurement error of the camera. By analyzing the displacement of the fixed target point, it is possible to monitor changes in the perspective projection matrix and thus compensate for camera measurement errors. In compensation experiments, using target displacement effectively predicts pixel drift in the pixel coordinate system. After compensation, the pixel error was suppressed from 1.99 pixels to 0.393 pixels. Repetitive measurement tests of the deep hole measurement instrument validate the practicality and reliability of compensating for thermal-induced errors using perspective transformation.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Improving Measurement Accuracy of Deep Hole Measurement Instruments through Perspective Transformation

Zhao,

Du,

2024

Sensors

View full text Add to dashboard Cite

show abstract

“…In this respect, the authors of [14] present a thermal camera that allows the adjustment of different factors concerning the temperature and ambient humidity. However, they point out that these have relatively little effect in the case of close-range applications, with building facades not exceeding 26 m in height and continuous facade lengths of up to 100 m. The paper by Zhu et al [15] comprises a study concerning the thermal inaccuracies that occur in cameras that are used to perform videogrammetry. A plane displacement function is proposed in order to adjust the temperatures measured, and several tests are subsequently carried out with different orientations of the camera so as to avoid the thermal drift caused by movement.…”

Section: Apparent Temperature In 3d Thermal Point Cloudsmentioning

confidence: 99%

Correction of Temperature from Infrared Cameras for More Precise As-Is 3D Thermal Models of Buildings

et al. 2023

View full text Add to dashboard Cite

This paper studies how to create precise 3D thermal maps of building interiors by correcting the raw apparent temperature values yielded by the thermal cameras. This matter has not, to date, been dealt with in the literature, which leads us to conclude that the current methodologies that obtain thermal point clouds are incomplete and imprecise. Without a suitable correction, any further energy parameter calculation obtained from the thermal point cloud is incorrect. This paper presents a method that deals with important aspects to be corrected, such as the repeatability of thermal cameras, the use of the true emissivity of the materials sensed, and the inclusion of the reflected radiant energy caused by the environment. The method has been successfully tested in several indoor scenes using a thermal scanning platform. The results show that significant corrections of up to 8% of the raw temperature values must be carried out in the final thermal model, thus justifying the need for the correction. As an application of the method, an empirical calculation and a comparison of transmittances with and without temperature corrections are presented at the end of the paper. In this case, the relative errors with respect to the average nominal U-value decrease from 94% to 11%. The general conclusion is that precise calculations of energy parameters in which infrared cameras are involved must take these corrections to temperature into account.

show abstract

“…However, existing manual methods suffer from the disadvantages of unreliable inspection results and considerable time consumption 1–4 . With the rapid development of deep learning (DL) and the establishment of benchmark datasets, vision‐based structural damage detection has garnered increasing attention 5–9 . Crack detection and identification is a common task that applies DL to a specific type of damage detection.…”

Section: Introductionmentioning

confidence: 99%

“…[1][2][3][4] With the rapid development of deep learning (DL) and the establishment of benchmark datasets, vision-based structural damage detection has garnered increasing attention. [5][6][7][8][9] Crack detection and identification is a common task that applies DL to a specific type of damage detection. For instance, Zhang et al 10 proposed a method based on a convolutional neural network (CNN) to detect road cracks, which exhibited superior results in comparison to manual feature extraction.…”

mentioning

confidence: 99%

Continual‐learning‐based framework for structural damage recognition

Shu

Ding

Zhang

et al. 2022

Structural Contr & Hlth

View full text Add to dashboard Cite

Existing convolutional neural network (CNN)-based methods have limitations in long-term multi-damage recognition for civil infrastructures. Owing to catastrophic forgetting, the recognition accuracy of such networks decreases when structural damage types keep increasing progressively, not to mention other issues such as an increased number of model parameters and data storage.Thus, this study proposes a continual-learning-based damage recognition model (CLDRM) for the recognition of multi-damage and relevant structural components in civil infrastructures. By integrating the Learning without Forgetting (LwF) method into the residual network with 34 layers, the CLDRM can be continuously trained for multiple recognition tasks without using the data from old tasks. The performance of the CLDRM is experimentally validated through four recognition tasks, namely, damage level, spalling check, component-type determination, and damage-type determination, and it is compared to the performance of a conventional CNN with feature extraction, fine-tuning, duplication and fine-tuning, and joint training, respectively. In addition, the effects of changes in three parameters, namely, distillation temperature, feature correlation between tasks, and learning order, are further investigated to explore the optimal model parameters and applicable scenarios in multi-damage recognition. CLDRM gradually aggregates the features of continuous tasks through knowledge distillation, which provides higher recognition accuracy for both old and new tasks while maintaining the advantages of computational cost and data storage. The research outcome is expected to meet the long-term requirements of handling progressively increasing multi-type damage recognition tasks for civil infrastructures.

show abstract

Mitigating thermal‐induced image drift for videogrammetric technique in support of structural monitoring applications

Cited by 3 publications

References 44 publications

Improving Measurement Accuracy of Deep Hole Measurement Instruments through Perspective Transformation

Improving Measurement Accuracy of Deep Hole Measurement Instruments through Perspective Transformation

Correction of Temperature from Infrared Cameras for More Precise As-Is 3D Thermal Models of Buildings

Continual‐learning‐based framework for structural damage recognition

Contact Info

Product

Resources

About