A machine vision system for detecting apples in orchards was developed. The system was designed to be used in harvesting robots and is based on a YOLOv3 algorithm with special pre- and post-processing. The proposed pre- and post-processing techniques made it possible to adapt the YOLOv3 algorithm to be used in an apple-harvesting robot machine vision system, providing an average apple detection time of 19 ms with a share of objects being mistaken for apples at 7.8% and a share of unrecognized apples at 9.2%. Both the average detection time and error rates are less than in all known similar systems. The system can operate not only in apple-harvesting robots but also in orange-harvesting robots.

Despite the great possibilities of modern neural network architectures concerning the problems of object detection and recognition, the output of such models is the local (pixel) coordinates of objects bounding boxes in the image and their predicted classes. However, in several practical tasks, it is necessary to obtain more complete information about the object from the image. In particular, for robotic apple picking, it is necessary to clearly understand where and how much to move the grabber. To determine the real position of the apple relative to the source of image registration, it is proposed to use the Intel Real Sense depth camera and aggregate information from its depth and brightness channels. The apples detection is carried out using the YOLOv3 architecture; then, based on the distance to the object and its localization in the image, the relative distances are calculated for all coordinates. In this case, to determine the coordinates of apples, a transition to a symmetric coordinate system takes place by means of simple linear transformations. Estimating the position in a symmetric coordinate system allows estimating not only the magnitude of the shift but also the location of the object relative to the camera. The proposed approach makes it possible to obtain position estimates with high accuracy. The approximate root mean square error is 7–12 mm, depending on the range and axis. As for precision and recall metrics, the first is 100% and the second is 90%.

Lifescience Global - Canada based international publisher for research professionals worldwide

Fintech is today not only a hot mass media discussion of the future of the financial sector, but also real projects that change banking and financial services. The paper describes features and characteristics of contemporary Russian fintech landscape and ecosystem. The examples of innovative financial services in Russia, including online banking and accounting, new payments and transfers services, platforms for crowdfunding and peer-to-peer lending, blockchain initiatives, etc. are discussed. It is shown that fintech initiatives have not yet led to a radical transformation of the financial sector in Russia because participants of the fintech ecosystem have different points of view on fintech. Russian banks are now developing fintech initiatives within themselves, encouraging technology companies and fintech startups to focus their efforts on innovations that are aimed at improving processes, rather than opening new markets. The Government directs the main efforts to initiatives related to regulation of cryptocurrencies circulation and to introduction of blockchain in regtech and cybersecurity. Customers are interested in new and more convenient functionality in mobile applications, and they are waiting for new value propositions, including fast international money transfers, roboadvising, personal financial management, peer-to-peer lending.

Forecasting financial market trends is one of the most important tasks for investors. Traditional methods are complemented by machine learning methods, the potential of which is to find hidden knowledge in large amounts of information. Nonlinear models and methods based on artificial intelligence (neural networks) are considered. Currently, the real use of neural networks is developing in the following areas: stock exchange and macroeconomic forecasting; speech recognition and dialogue with the person; imitation of intellectual activity; improvement of poor-quality and noisy information; identification of suspicious persons and situations. In this paper it will be shown about the effective use of neural networks in forecasting problems.