I. Introduction
Digital transformation in all areas of life is a factor for effective implementation of innovative information and computer technologies in precision agricultural production as well. The massive use of smart information and communication systems requires the development of mobile, web-based and cloud-based technologies that are accessible and easily applicable to the modern farmer [1]. Modern smart agronomic technologies for growing agricultural crops increasingly require fast and timely information from the field and transmission to mobile devices as well. Therefore, it is necessary to develop and implement methods for express, monitoring of the composition and properties of the soil, which will contribute to finding the much-needed overall balance between preserving the environment, natural resources and meeting the growing needs of agricultural products. The application of sustainable smart agriculture and the maintenance of the ecological condition of the soils is especially imperative under the conditions of the significant climate changes reported in recent years. The express and timely monitoring of the soil implies the determination of a large number of indicators - macro and microelements, soil pH, humus, humidity at each stage of the phenological development of the cultivated crops. In traditional laboratory measurements, soil samples are examined with analytical methods that include combustion, oxidation, use of chemical substances and harmful products [3]. This type of laboratory measurement is time consuming and requires trained personnel. These are prerequisites to look for innovative, express methods for measuring soil quality directly in the field. Known diagnostic tools for rapid assessment of soil quality are based on spectral analysis in the UV-VIS-NIR region; satellite images, unmanned aerial systems UAV and others [4], [5]. However, non-contact, non-destructive and more accessible methods for soil analysis and evaluation directly in the field have not yet been proposed. Soil color is considered to be easy to measure and can provide a quick assessment of some basic soil properties [2]. Obtaining visual information about the agrochemical parameters of the cultivated field does not require complex sample preparation [7], helps to make timely decisions about the soil condition before important agrotechnical stages or diseases have occurred in the cultivated crops [3]. In the literature, there are a number of examples of finding connections between color and soil composition, such as the presence of moisture, soil organic matter, iron impurities, etc. [3], [8], but despite the progress in this field, there are still aspects not covered and opportunities to search for effective and rational solutions for remote determination of soil quality by its color. For the conditions in Bulgaria, similar methods for express monitoring of agrochemical indicators of the soil have not been applied. All this gives reason to undertake a scientific study on the possibility of analyzing the color characteristics of digital soil images as an alternative and fast method for determining basic soil indicators.