Improving a neural network model for semantic segmentation of images of monitored objects in aerial photographs

Abstract

This paper considers a model of the neural network for semantically segmenting the images of monitored objects on aerial photographs. Unmanned aerial vehicles monitor objects by analyzing (processing) aerial photographs and video streams. The results of aerial photography are processed by the operator in a manual mode; however, there are objective difficulties associated with the operator's handling a large number of aerial photographs, which is why it is advisable to automate this process. Analysis of the models showed that to perform the task of semantic segmentation of images of monitored objects on aerial photographs, the U-Net model (Germany), which is a convolutional neural network, is most suitable as a basic model. This model has been improved by using a wavelet layer and the optimal values of the model training parameters: speed (step) ‒ 0.001, the number of epochs ‒ 60, the optimization algorithm ‒ Adam. The training was conducted by a set of segmented images acquired from aerial photographs (with a resolution of 6,000×4,000 pixels) by the Image Labeler software in the mathematical programming environment MATLAB R2020b (USA). As a result, a new model for semantically segmenting the images of monitored objects on aerial photographs with the proposed name U-NetWavelet was built. The effectiveness of the improved model was investigated using an example of processing 80 aerial photographs. The accuracy, sensitivity, and segmentation error were selected as the main indicators of the model's efficiency. The use of a modified wavelet layer has made it possible to adapt the size of an aerial photograph to the parameters of the input layer of the neural network, to improve the efficiency of image segmentation in aerial photographs; the application of a convolutional neural network has allowed this process to be automatic.

Improving the model of object detection on aerial photographs and video in unmanned aerial systems

“…To advance the proposed model, it is planned: -to increase the base of marked aerial photographs for a training sample [20,21];…”

Some Aspects of Artificial Intelligence Development Strategy for Mobile Technologies

“…It's easy to predict that such a service will be in demand in smartphones shortly. It's entirely possible that this approach will gradually replace traditional image segmentation technologies [11] and object detection [12,13], although there will still be a niche for them in extremely resource-limited mobile applications. Moreover, there may be some integration of traditional neural network architectures with transformer structures.…”

Modern Materials for Fire Protection of Reinforced Concrete Agro-Industrial Structures

“…It is especially urgent for the constructions of the agroindustrial complex in the south of Ukraine, including the Mykolayiv region. Given that the agricultural structures are located at considerable distances from each other, which greatly complicates the delivery of firefighting equipment, important, in our view, is the monitoring by drones of the centers of destruction and fires with modern means of video surveillance and fixation [1,2]. It is reasonable to make the processing of received information with help of worked out method of wavelet-transformation [3] and recognition models based on neural networks [4], a choice of modern means of passive and active fire protection of undamaged structures of agroindustrial complex is made.…”