Deep Learning Based Object Recognition in Real Time Images Using Thermal Imaging System

An efficient driver assistance system is essential to avoid mishaps. The collision between the vehicles and objects before vehicle is the one of the principle reason of mishaps that outcomes in terms of diminished safety and higher monetary loss. Researchers are interminably attempting to upgrade the safety means for diminishing the mishap rates. This paper proposes an accurate and proficient technique for identifying objects in front of vehicles utilizing thermal imaging framework. For this purpose, image dataset is obtained with the help of a night vision IR camera. This strategy presents deep network based procedure for recognition of objects in thermal images. The deep network gives the model understanding of real world objects and empowers the object recognition. The real time thermal image database is utilized for the training and validation of deep network. In this work, Faster R-CNN is used to adequately identify objects in real time thermal images. This work can be an incredible help for driver assistance framework. The outcomes exhibits that the proposed work assists to boost public safety with good accuracy.

Non-Contact Spirometry Using a Mobile Thermal Camera and AI Regression

Non-contact physiological measurements have been under investigation for many years, and among these measurements is non-contact spirometry, which could provide acute and chronic pulmonary disease monitoring and diagnosis. This work presents a feasibility study for non-contact spirometry measurements using a mobile thermal imaging system. Thermal images were acquired from 19 subjects for measuring the respiration rate and the volume of inhaled and exhaled air. A mobile application was built to measure the respiration rate and export the respiration signal to a personal computer. The mobile application acquired thermal video images at a rate of nine frames/second and the OpenCV library was used for localization of the area of interest (nose and mouth). Artificial intelligence regressors were used to predict the inhalation and exhalation air volume. Several regressors were tested and four of them showed excellent performance: random forest, adaptive boosting, gradient boosting, and decision trees. The latter showed the best regression results, with an R-square value of 0.9998 and a mean square error of 0.0023. The results of this study showed that non-contact spirometry based on a thermal imaging system is feasible and provides all the basic measurements that the conventional spirometers support.

Reducing inter-operator variability when measuring subcutaneous tumours in mice: An investigation into the effect of inter-operator variability when using callipers or a novel 3D and thermal measurement system

Consistent, repeatable tumour volume measurements are key to accurately calculating tumour growth and successful assessment of therapeutic efficacy. Our previous work showed that a novel 3D and thermal imaging system for measuring subcutaneous rodent tumours (BioVolume) significantly reduced inter-operator variability across three in vivo efficacy studies. Here we continue to investigate this reduction in inter-operator variability across a much larger dataset. A dataset of 5,593 inter-operator repeats and 5,073 corresponding calliper repeats was obtained from tumour scans and measurements in 22 laboratories across 238 studies with 112 users, 23 animal strains and 98 unique cell lines. The inter-operator variability of the two measurement methods was analysed using coefficient of variation (CoV), intra-class correlation (ICC) analysis, and significance testing. The 3D and thermal imaging system produced a significantly lower median CoV of 0.173 compared to a median calliper CoV of 0.205 (P value = 5.2 x 10^-9). ICC analysis further confirmed the statistical significance of these values, allowing us to conclude that this novel 3D and thermal imaging system offers a significant reduction in inter-operator variability. This has the potential to improve reproducibility of in vivo studies across a wide range of animal strains and cell lines. The effects of using a device with large inter-operator variability at critical points in the study were also investigated. At randomisation, changing the operator performing measurements resulted in 59.4% probability that a rodent would be reassigned to a different group. For measurements carried out using the imaging system, the probability that changing the operator would also result in change of a rodent’s group was much lower at 29.2%. During studies where the tumour was expected to regress, substituting an operator mid-study resulted in a tumour volume increase of approximately 500mm^3 when callipers were used for measurement. For the imaging device, substituting users did not affect the tumour regression trend, potentially removing the need for the same operator to be present for the entire study duration. The effect of swapping an operator mid-study on the drug efficacy metric AUC (Area Under the Curve) was also observed; no statistical difference in AUC was observed for both BioVolume and callipers (overlapping 95% confidence intervals).

Design of a Thermal Imaging System to Detect Possible Cases of COVID-19 Patients

Abstract— The problem that the world is currently facing and that has claimed more than 3.2 million lives worldwide is COVID-19. Being a highly contagious disease, the WHO recommended limiting the movement of people out of their residence. Given this, Peru took the necessary measures to control the spread of this virus, therefore, the government decreed a general quarantine in the country, which from March 6th to May 9th, 2020, had reported the death of 1814 people in the country, because the health system at the national level was not prepared for such magnitude, in addition to that the number of infections continued to grow since they do not respect social distancing. In view of this problem, this article will design a thermal image processing system to detect possible cases of patients with COVID-19, in such a way that it allows companies or institutions to know the body temperature of each person, and thus know they are possibly COVID-19 patients. Through the design of this system, it will be possible to measure body temperature with the drone and the thermal camera at 50 cm from the person, in such a way that if the person has a body temperature higher than 38°C it could be infected with COVID-19. Therefore, the implementation of this system will help reduce the number of infections within an institution or workplace. Keywords-- Thermal camera, COVID-19, Drone, MATLAB, Image.

Application of Zernike moments in computer vision problems for infrared images

The efficiency of using Zernike moments when working with digital images obtained in the infrared region of the spectrum is considered to improve the accuracy and speed of an autonomous thermal imaging system. The theoretical justification of the choice of Zernike moments for solving computer (machine) vision problems and the choice of a suitable threshold binarization method is given. In order to verify the adequacy and expediency of using the chosen method, practical studies were conducted on the use of Zernike methods for distorting various thermal images in shades of gray.

Application of thermal imaging for detecting cold air leak location in cold storage

<span>Nowadays Cold storage has been playing an important role in preserving the perishable food products like fruits, vegetables, dairy products, fish, and meat etc. The major problem of cold storage is unwanted energy transmission during the operation. It is necessary to maintain the constant temperature during storing the products. There are different kinds of energy loss happening during the operation. One of the major causes is cold air leaking from the inside of the cold room to outside. It is due to poor insulation and improper maintenance of cold storage. It is very difficult to identify the exact location of the leak by the naked eye. In this research work, the cold storage was inspected with the thermal imaging system. Thermal cameras are highly sensitive to temperature and it can detect the variation of temperature ranging from 0.1°C. The measured temperatures are further converted into a colour based pattern. It is known as thermogram. These colour-based thermal patterns are further processed for identifying energy transmission location. It is done by applying various image processing methods such as histogram equalization, diffusion error, otsu thresholding and morphologic function. These techniques were applied to images of cold storages and exact cold air transmission locations were identified.</span>