Search

Purpose. Investigation of the possibilities of increasing the accuracy and variability of recognition of different groups of objects with similar redundant information by a neural network, after preprocessing and grouping of training images for further use on microcontrollers. The effect of removing redundant information in the training data on the practical values of recognition accuracy of different categories of objects based on the MobileNet V2 architecture is verified. Methodology. Creating different groups of training images using the Edge Impulse software platform. Image processing by the method of removing unnecessary objects in the graphic editor Adobe Photoshop. Testing the recognition accuracy of raw, processed and mixed groups of training images. Findings. Several groups of training images have been created on the basis of the Edge Impulse software platform. In part of the images, all unnecessary elements were removed using the Adobe Photoshop graphic editor. It was established that in the presence of similar redundant information on the training data, in different categories of objects, unrecognized and falsely recognized test images appear as a result of recognition. It was experimentally found that the method of removing redundant information from training images gives a clearer distribution of features, and the combination of raw and processed training data gives an average increase in recognition accuracy of more than 10% for each category, as well as a significant reduction of unrecognized and incorrectly recognized test images, while the same amount of training data. Originality. A combined method of training data processing and grouping was obtained, which increases the accuracy of object recognition without increasing the amount of training data. The effect of similar redundant information in different categories of objects on recognition accuracy was investigated. Practical value. The obtained results make it possible to increase the accuracy of recognition of different groups of objects with similar redundant information by one neural network without increasing the number of training images. [ABSTRACT FROM AUTHOR]

Purpose. Research on the dependence of the amount of microcontroller memory used and the image recognition time on the type of convolutional neural network MobileNet V1 or V2 and their hyperparameters. Methodology. Creating a database of images necessary for training a neural network using the Edge Impulse software platform with subsequent loading of the network into the memory of a 32-bit microcontroller ESP32 for practical evaluation of network characteristics. Findings. A comparison of the characteristics of MobileNetV1 and MobileNetV2 neural networks was made. Experiments were conducted to determine the dependence of the recognition time of selected objects on the image, the amount of used RAM and program memory, based on the ESP-EYE microcontroller with a camera, depending on the network width ratio, image size and convolutional neural architecture network. It was determined that the time of image classification for the MobileNetV2 network model takes from three to ten seconds or more, which is not acceptable for fast recognition tasks. It was also found that the microcontroller program memory is insufficient to analyze the 160 by 160 pixel images with the maximum network width of the MobileNetV2 model. Using the MobileNetV1 network provides slightly lower recognition accuracy, but requires significantly less microcontroller resources and time. Originality. The peculiarities of the use, possibilities and limitations of neural networks for image recognition in systems based on ESP32 microcontrollers have been verified in practice. The dependence of the amount of memory used on the microcontroller and the time of image recognition on the type of convolutional neuron was established in order to select additional image processing tools to improve the quality of recognition. Practical value. The obtained results allow the proper selection of the MobileNetV1 or MobileNetV2 neural network depending on the specific tasks of image recognition by systems on microcontrollers. [ABSTRACT FROM AUTHOR]