Denoising and fuel spray droplet detection from light-scattered images using deep learning

A deep learning-based method for denoising and detecting the gas turbine engine spray droplets in the lightscattered image(Mie scattering)is proposed for the first time.A modified U-Net architecture is employed in the proposed method to denoise and regenerate the droplets.We have compared and validated the performance of the modified U-Net architecture with standard conventional neural networks(CNN)and modified ResNet architectures for denoising spray images from the Mie scattering experiment.The modified U-Net architecture performed better than the other two networks with significantly lower Mean Squared Error(MSE)on the validation dataset.The modified U-Net architecture also produced images with the highest Power Signal to Noise Ratio(PSNR)compared to the other two networks.This superior performance of the modified U-Net architecture is attributed to the encoder-decoder structure.During downsampling,as part of the encoder,only the most prominent features of the image are selectively retained by excluding any noise.This reconstruction of the noisefree features has produced a more accurate and better denoised image.The denoised images are then passed through a center predictor CNN to determine the location of the droplets with an average error of 1.4 pixels.The trained deep learning method for denoising and droplet center detection takes about 2.13 s on a single graphics processing unit(GPU).This study shows the promise for real-time processing of the experimental data using the well-optimized network.

Development and Evaluation of an Optical Sensing System for Detection of Herbicide Spray Droplets

Real time monitoring of herbicide spray droplet drift is important for crop production management and environmental protection. Existing spray droplet drift detection methods, such as water-sensitive paper and tracers of fluorescence and Rubidium chloride, are time-consuming and laborious, and the accuracies are not high in general. Also, the tracer methods indirectly quantify the spray deposition from the concentration of the tracer and may change the drift characteristics of the sprayed herbicides. In this study, a new optical sensor system was developed to directly detect the spray droplets without the need to add any tracer in the spray liquid. The system was prototyped using a single broadband programmable LED light source and a near infrared sensor containing 6 broadband spectral detectors at 610, 680, 730, 760, 810, and 860 nm to build a detection system for monitoring and analysis of herbicide spray droplet drift. A rotatory structure driven by a stepper motor in the system was created to shift the droplet capture line going under the optical sensor to measure and collect the spectral signals that reflect spray drift droplets along the line. The system prototype was tested for detection of small (Very Fine and Fine), medium (Medium), and large (Coarse) droplets within the droplet classifications of the American Society of Agricultural and Biological Engineers. Laboratory testing results indicated that the system could detect the droplets of different sizes and determine the droplet positions on the droplet capture line with 100% accuracy at the wavelength of 610 nm selected from the 6 bands to detect the droplets.