The article discusses the problem of detecting sick or mechanically damaged potatoes using machine learning methods. We proposed an algorithm and developed a system for the rapid detection of damaged tubers. The system can be installed on a conveyor belt in a vegetable store, and it consists of a laptop computer and an action camera, synchronized with a flashlight system. The algorithm consists of two phases. The first phase uses the Viola-Jones algorithm, applied to the filtered action camera image, so it aims to detect separate potato tubers on the conveyor belt. The second phase is the application of a method that we choose based on video capturing conditions. To isolate potatoes infected with certain types of diseases (dry rot, for example), we use the Scale Invariant Feature Transform (SIFT)—Support Vector Machine (SVM) method. In case of inconsistent or weak lighting, the histogram of oriented gradients (HOG)—Bag-of-Visual-Words (BOVW)—neural network (BPNN) method is used. Otherwise, Otsu’s threshold binarization—a convolutional neural network (CNN) method is used. The first phase’s result depends on the conveyor’s speed, the density of tubers on the conveyor, and the accuracy of the video system. With the optimal setting, the result reaches 97%. The second phase’s outcome depends on the method and varies from 80% to 97%. When evaluating the performance of the system, it was found that it allows to detect and classify up to 100 tubers in one second, which significantly exceeds the performance of most similar systems.

Detecting sugar beetroot crops with mechanical damage using machine learning methods is necessary for fine-tuning beet harvester units. The Agrifac HEXX TRAXX harvester with an installed computer vision system was investigated. A video camera (24 fps) was installed above the turbine, which receives the dug-out beets after the digger and is connected to a single-board computer. At the preprocessing stage, static and insignificant image details were revealed. Canny edge detector and excess green minus excess red (ExGR) method were used. The identified areas were excluded from the image. The remaining areas were glued with similar areas of another image. As a result, the number of images entering the second stage of preprocessing was reduced by half. Then Otsu’s binarization was used. The main stage of image processing is divided into two sub-stages: detection and classification. The improved YOLOv4-tiny method was chosen for root crop detection using a single-board computer (SBC). This method allows processing up to 14 images of 416 × 416 pixels with 86% precision and 91% recall. To classify root crop damage, we considered two algorithms as candidates: 1. bag of visual words (BoVW) with a support vector machine (SVM) classifier using histogram of oriented gradients (HOG) and scale-invariant feature transform (SIFT) descriptors; 2. convolutional neural networks (CNN). Under normal lighting conditions, CNN showed the best accuracy, which ranged from 97% to 100%, depending on the damage class. The implemented methods were used to detect and classify blurred images of sugar beetroots, which were previously rejected. For improved YOLOv4-tiny precision was 74% and recall was 70%. CNN classification accuracy ranged from 90% to 95% depending on the root crops damage class.

The article proposes a methodology for optimizing the process of irrigation of crops using a phytoindication system based on computer vision methods. We have proposed an algorithm and developed a system for obtaining a map of irrigation for maize in low latency mode. The system can be installed on a center pivot irrigation and consists of 8 IP cameras connected to a DVR connected to a laptop. The algorithm consists of three stages. Image preprocessing stage - applying an integrated excess green and excess red difference (ExGR) index. The classification stage is the application of the method that we choose depending on the system’s operating conditions. At the final stage, a neural network trained using the Resilient Propagation method is used, which determines the rate of watering of plants in the current sector of the location of the sprinkler. The selected methods of pretreatment and classification made it possible to achieve an accuracy of plant identification up to 93%, growth stages - up to 92% (with unconsolidated maize sowing and good lighting). System performance up to 100 plants in one second, which exceeds the performance of similar systems. The neural network showed an accuracy of 92% on the training set and 87% on the test set. Dynamic analysis of spatial and temporal variability leads to an increase in productivity and efficiency of water use. In addition, given the ubiquitous distribution of agribusiness management systems, this approach is quite simple to implement in the farm’s conditions.

The purpose of the article is to create an effective method to monitor the state of the drill string and the bit without interfering with the drilling process itself in low-time delay mode. For continuous monitoring of the well drilling process, an experimental setup was developed that operates on the basis of the use of the phase-metric method of control. Any movement of the bit causes a change in the electrical characteristics of the probing signal. To obtain a stable signal from a bit immersion depth of up to 250 m, a frequency of probing electrical signals of 166 Hz and an amplitude of up to 500 V were used; sampling rate (analog-to-digital converter) ADC - 10101 Hz. To identify the state of the drill string and the bit according to the graphs of dependences of changes in the electrical characteristics of the probing signal on time, the authors of the article investigated a number of deep learning methods, based on the results of the research, a line of capsule neural network (CapsNet) methods was selected. The authors have developed two modifications of 1D-CapsNet and Windowed Fourier Transform (WFT) - 2D-CapsNet. To identify the transition between two rock layers with different properties, WFT-2D-CapsNet showed an accuracy of 99%, which is 2-3% higher than the results of modern rock studies based on measurement-while-drilling (MWD) and logging-while-drilling (LWD) methods. The WFT-2D-CapsNet method unambiguously detects self-oscillations in the drill string and detects the good condition of the bit with an accuracy of 99%.

This article deals with the multicriteria programming model to optimize the time of completing home assignments by school students in both in-class and online forms of teaching. To develop a solution, we defined 12 criteria influencing the school exercises’ effectiveness. In this amount, five criteria describe exercises themselves and seven others the conditions at which the exercises are completed. We used these criteria to design a neural network, which output influences target function and the search for optimal values with three optimization techniques: backtracking search optimization algorithm (BSA), particle swarm optimization algorithm (PSO), and genetic algorithm (GA). We propose to represent the findings for the optimal time to complete homework as a Pareto set.