Double Thresholding with Sine Entropy for Thermal Image Segmentation

Traditional thresholding methods are often used for image segmentation of real images. However, due to distinct characteristics of infrared thermal images, it is difficult to ensure an optimal image segmentation using the traditional thresholding algorithms, and therefore, sometimes this can lead to over-segmentation, missing object information, and/or spurious responses in the output. To overcome these issues, we propose a new thresholding technique that makes use of the sine entropy-based criterion. Moreover, we build a double thresholding technique that makes use of two thresholds to get the final image thresholding result. Besides, we introduce the sine entropy concept as a supplement of the Shannon entropy in creating threshold-dependent criterion derived from the grayscale histogram. We found that the sine entropy is more robust in interpreting the strength of the long-range correlation in the gray levels compared to the Shannon entropy. We have experimented with our method on several infrared thermal images collected from standard image databases to describe the performance. On comparing with the state-of-art methods, the qualitative results from the experiments show that the proposed method achieves the best performance with an average sensitivity of 0.98 and an average misclassification error of 0.01, and second-best performance with an average sensitivity of 0.99 and an average specificity of 0.93.

Deep Learning-Based Thermal Image Analysis for Pavement Defect Detection and Classification Considering Complex Pavement Conditions

Automatic damage detection using deep learning warrants an extensive data source that captures complex pavement conditions. This paper proposes a thermal-RGB fusion image-based pavement damage detection model, wherein the fused RGB-thermal image is formed through multi-source sensor information to achieve fast and accurate defect detection including complex pavement conditions. The proposed method uses pre-trained EfficientNet B4 as the backbone architecture and generates an argument dataset (containing non-uniform illumination, camera noise, and scales of thermal images too) to achieve high pavement damage detection accuracy. This paper tests separately the performance of different input data (RGB, thermal, MSX, and fused image) to test the influence of input data and network on the detection results. The results proved that the fused image’s damage detection accuracy can be as high as 98.34% and by using the dataset after augmentation, the detection model deems to be more stable to achieve 98.35% precision, 98.34% recall, and 98.34% F1-score.

Adaptive Thermal Image Velocimetry of spatial wind movement on landscapes using near target infrared cameras

Thermal Image Velocimetry (TIV) is a near-target remote sensing technique for estimating two- dimensional near-surface wind velocity based on spatiotemporal displacement of fluctuations in surface brightness temperature captured by an infrared camera. The addition of an automated parameterization and the combination of ensemble TIV results into one output made the method more suitable to changing meteorological conditions and less sensitive to noise stemming from the airborne sensor platform. Three field campaigns were carried out to evaluate the algorithm over turf, dry grass and wheat stubble. The derived velocities were validated with independently acquired observations from fine wire thermocouples and sonic anemometers. It was found that the TIV technique correctly derives atmospheric flow patterns close to the ground. Moreover, the modified method resolves wind speed statistics close to the surface at a higher resolution than the traditional measurement methods. Adaptive Thermal Image Velocimetry (A-TIV) is capable of providing contact-less spatial information about near-surface atmospheric motion and can help to be a useful tool in researching turbulent transport processes close to the ground.

Live Human Detection Robot in Earthquake Conditions

This report outlines a human searching device that takes the form of a robotic car and serves as a backup mechanism for saving lives in the event of a disaster. The temperature sensor, in general, detects the thermal image of the human body, and there has been extensive research into human searching with the gas and humidity sensor. In the intelligent robot device’s study, achieving accurate and reliable human detection and tracking is a difficult challenge. The architecture of human detection and tracking mechanisms over non-overlapping field of views is examined in this paper. To compensate for their respective flaws, a search method is proposed. The proposed method’s rate and accuracy of human detection was tested in an experimental setting. We may guide the robot’s movement by commanding it to move left, right, forward, or backward. We plan to equip the robot with sensors that will enable us to track and detect humans behind the wall.