work is devoted to the development of a method for accounting for the projective cover of plants based on the use of photographic sites. The assessment of the counting lawn areas showed the high accuracy of the developed accounting system.
Introduction
Today, an acute problem of modern urban areas is the rapid deterioration of the ecological situation. Among the many existing environmental problems of modern cities, special attention is paid to the reduction of areas of green spaces. Due to the excessive content of pollutants in the environment and the low fertility of urban soils, the degradation of green spaces occurs. Thus, it becomes necessary to restore the state and increase the number of green spaces in the city [1].
In recent years, along with shrubs and trees and shrubs, various types of lawn herbage are increasingly used in urban landscaping. The use of lawns for landscaping has a number of significant advantages: high growth rate, resistance to mechanical damage, a stable root system, unpretentiousness and high decorative effect [2].
When assessing the quality of lawn grass stands, the most often used methodology developed by A. Laptev [3]. This method is complex and quite time consuming. Lawn grasses are assessed in terms of shoot productivity (on a 6-point scale), as well as general decorativeness (on a 5-point scale).
Shoot productivity is determined by counting the number of shoots per 1 m2. For this, the number of shoots in a certain area is counted (most often 10x10 cm), after a series of such measurements, the average value is displayed and converted to 1 m 2 .
The projective grass cover of the soil is determined visually, looking from top to bottom at an angle of 90Λ to the grass. Determine what part of the area is covered with grass and express this value as a percentage.
During field studies, this process takes a long period of time, however, some authors indicate that the definition of the projective cover can be carried out using computer transformations of a color image of the level in a horizontal projection [4].
The use of information technologies for assessing the projective cover of geobotanical layers can reduce the time and labor costs for the assessment process, as well as increase the accuracy of measurements [5].
We have proposed a method for determining the projective coverage of lawn grass stands, based on the automatic selection of the contours of plants in digital images.
In this regard, the aim of the work is to assess the quality of herbage existing in Alchevsk using the LawnMaster system developed by us.
Research objectives
- Evaluation of productivity indicators of shoots and projective cover of lawn grass stands in Alchevsk.
- Visual determination of the projective coverage of the counting areas of lawn herbage.
- Determination of the projective coverage of the counting areas of lawn herbage using the LawnMaster system.
- LawnMaster .
The projective cover and the productivity of shoot formation of the counting areas of lawn herbage, which were formed on the territory of building No. 6 of DonSTU, were evaluated. On the basis of the indicators of the projective cover and the productivity of shoot formation, a general indicator of the quality of herbage was derived [3].
Shoot productivity was determined by counting shoots on an area of ββ10x10 cm, after which the resulting indicator was converted to 1 m2. Shoot formation productivity was measured with a five-fold repetition, and the average was derived from the obtained indicators.
The projective cover of lawn herbage was determined using the LawnMaster system.
To create the program, the Python programming language and the OpenCV library were used.
OpenCV provides a source code library including open source for image processing.
The resulting image is converted from bgr to lab format, since it is easier to select the border of the lawn coverage in this format.
After that, a mask is created for this image based on the selected coefficients that
determine the allowed and forbidden areas of the image. These coefficients were selected using trackbars, which were created based on the data of the same library.
After creating a mask, the percentage of allowed pixels is calculated on it, which is displayed on the image. For more convenient use of the program, a simple command line was created, in which you can enable / disable the mask, process another image and exit the program.
To test the effectiveness of the created program, we used photographs of lawn grass stands from the experimental plots of the Department of Ecology and Railway Transport of DonSTU. The survey sites were photographed from a height of 1 m at an angle of 90Λ. Also, the grass stands were visually assessed the projective cover (by a group of three people, from the readings of which the average was derived), after which the data were compared with those obtained after image processing in the LawnMaster system.
Figures 2,3,4,5 show photographs of the surveyed counting areas of parterre lawn herbage.
Figure 1 Accounting site # 1
Figure 2 Accounting site # 2
Figure 3 Accounting site # 3
Figure 4 Accounting site # 4
Research results
The results of assessing the quality of lawn grass stands located on the territory of building No. 6 of DonSTU are shown in Table 1.
During the assessment, it was revealed that there are 3 groups of herbage of the highest quality, 4 groups of grass stands of excellent quality on the territory of the building. Grass stands of good to mediocre quality were also observed.
The results of the study of the projective coverage of the counting areas of lawn grass stands are shown in Table 1.
As can be seen from the table above, the LawnMaster system gives an estimate of the projective cover of lawn grass stands with a difference of 0.3-9% compared to the visual assessment.
conclusions
- Assessment of the decorativeness of lawn grass stands located on the territory of DonSTU showed that the studied grass stands belong to the group of grass stands of mediocre, good, excellent and superior quality.
- LawnMaster .
- (0,3-9%) LawnMaster , .
1. Adonyeva TB, Ivanova EM, Kalyuzhnaya LA Green spaces of the city of Voronezh: current state, problems // Vestnik VSU. - 2001. - P. 139.
2. Gladov A. V. Greening as a factor in improving the improvement of the city (on the example of the urban district of Samara) // Bulletin of the Samara State University. - 2015. - No. 2 (124).
3. Laptev A. A. Lawns // Kiev: Naukova Dumka. - 1983. - T. 243. - P. 4.
4. Balalaev A. K., Skripnik O. A. Preliminary results of the application of the method of digital image processing to determine the projective cover of vegetation as the main indicator of the state of ecosystems // Ecology and nature conservation. - 2011.
5. Buzuk GN, Sozinov OV Methods of accounting for the projective cover of plants: a comparative assessment with the use of photographic areas // News of the Samara Scientific Center of the Russian Academy of Sciences. - 2014. - T. 16. - No. 5-5