CC BY
20
Strip tillage and sowing fitomeliorants at the combination S25+D50+G25 (Shrabs - 25%, Dwarf shrubs - 50%, Grasses - 25%) creates favorable conditions for seeded fitomeliorants and allows to create all-year-round pastures. Total forage productivity restored/improved rangelands considerably more then other combinations of components.
Acknowledgments
This study was funded by the Centre of Science and Technology of Republic of Uzbekistan.
References:
1. Zhao H. L. at all., desertification process due to heavy grazing in sandy rangeland, Inner Mongolia. Journal of Arid Environments, -2005. - 62, 309-319.
2. Kassas M., Arid and Semi-Arid Lands: Problems and Prospects, Agro-Ecosystems, - 19773, - 185-204.
3. Loraine Van den Berg, Klaus Kellner, Restoring degraded patches in a semi-arid rangeland of South Africa. Journal of Arid Environments - 2005. - 61, 497-511.
4. Steven L. Rethinking Desertification: What Do We Know and What Have We Learned?//World Development. - 1991. - Vol. 19. - No 9. - P. 1137-1143.
5. Su Yong-Zhong at all., Influences of continuous grazing and livestock exclusion on soil properties in a degraded sandy grassland, Inner Mongolia, northern China., Catena, - 2005. - 59, 267-278.
6. Tatyana A. Saiko, Igor S. Zonn, Irrigation expansion and dynamic of desertification in the Circum-Aral region of Central Asia. Applied Geography, - 2000. - 20, 349-367.
7. Thomas D. S. G., Middleton N., Salinization: new perspectives on a major desertification issue. Journal ofA. Envir., - 1993. - 24, 95-105.
8. Ed Fredrickson at all, Perspectives on desertification: south-western United States., Journal ofArid Environments, - 1998. - 39, 191-207.
9. Gregory S. Okin at all, Journal ofArid Environments, - 2001. - 47, 123-144.
10. Gintiburger G., Toderich K. N., Mardonov B. K. and Mahvudov M. M., Ronglands of the arid and semi-arid zones in Uzbekistan., CIRAD-UCARDA., - 2003. - 426 p.
DOI: http://dx.doi.org/10.20534/ESR-17-1.2-189-192
Kamilov Mirzayan, Nosirov Khabibullo, Tashkent University of Information Technologies, E-mail: n.khabibullo1990@gmail.com
Brightness transformation of image with adaptive blocks, research of its efficiency in compression and estimation of reconstructed image quality
Abstract: This article discusses the method of converting the luminance of the adaptive block partitioning, which allows you to select and remove the image uniformity. It also provides the experimental results obtained by the study of the effectiveness of this method and the evaluation of the reconstructed images.
Keywords: image compression, conversion, block partition.
Introduction. Today there are many different techniques and video standards that provide good image quality at bit rate more than 3-5 Mbps. However, at the moment an important task is to develop efficient methods of compression, providing the transmission of broadcast TV programming stream via cellular communication at a speed of 2 Mbps [1].
Main part. To improve the efficiency of image compression the image's pixels' brightness changing the image has been proposed [2; 3]. The method starts work with dividing the image into square blocks of arbitrary size ranging from 32x32 to 2x2 pixels within which the luminance value is set in such a way that on the one hand, to bring it closer to zero, on the other hand, to make more uniform the entire brightness frame field, as shown in Figure 1.
As can be seen from Figure 1, on certain types of images can be obtained a full uniform image, which statistical information is well below 300-500 times by the RLE long series compressor.
However, these are provided only for certain types of synthetic images and on a normal picture some violation of the uni-
formity of brightness of the adjacent blocks (fig. 2) appears, which significantly reduces the effectiveness of the method. Therefore, we proposed adaptively changing block sizes method, depending on changes in the brightness of the source image structure [3].
In this algorithm, whose block diagram is shown in fig. 3, the procedure of finding the same areas of the image is carried out. This means that for the transmission of information on the areas having the same chrominance levels (monochrome), only the necessary area and color layer will transfer. This method is similar to the Huffman conversion, where the transmission of the same portion of the repeating elements, you must have one of the elements and the number of repetitions it [4].
This algorithm works as follows. The image is divided into blocks of size 2x2; between these 4 pixels the minimum luminance value will be found. Then, the minimum value is subtracted from each pixel in the block. Results are compared with the error factor, the size of which is determined by the formula 2,55*k (where k—is the user set percentage and is called the percentage error). If the difference is less than a pre-
determined number, then all the 4 pixels are considered to be similar, and will continue to be considered for the 2x2 block, and, in the further processing of the image of the four pixels of the block will be taken
away its minimum value, making it more uniform. If the difference is greater, respectively, the pixels are not similar, so will be written to the output file unchanged.
Fig.1. Original, converted and restored image colored stripes accordingly
Fig.2. Initial and compensated and restored the image with the appearance of the 'block' effect of the reconstruction accordingly
Figure 3. The block diagram of the adaptive partitioning the image on the variable-sized blocks
If the 2x2 blocks are found, they are compared with each other as well as pixels, i. e. four 2x2 block to be merged into a single block
of 4x4 pixels. This is done as follows: there is a minimum, i. e. the minimum value of 2x2 blocks (four) this is the minimum value for
a 4x4 block having 16 pixels. The obtained value is subtracted from the others the results are compared with factor error. If all four less than the difference, then the block will have a 4x4 size, if at least one value greater than a predetermined error, the 2x2 blocks are stored in separate blocks in a metafile. In this way, all the pixels are processed images, conventionally dividing into blocks. Comparing the blocks in the algorithm can be adjusted to the size of 32x32 pixel macroblock, that is, from the 1024 unit will be taken away by one (minimum) value.
After finding and conditional division into blocks of the image is processed, all found values are subtracted from the original image, depending on their location. The result is an image with the same dimensions, but changes the color. As greater identical areas in the image, as image is more uniform and thus the output can be highly compressed without deteriorating the visual quality. The block coefficients are written metadata signifying their size, and color of a
Table 1. - The plot of coi
minimum value so that the image can be restored in the decoding. For example, if the decoder reads from the array 5 metadata value (identifier 32x32 block), it means that a value of a 32x32 pixel block in the need to add the next number is the minimum value for this block.
To evaluate the efficiency of this method a number of studies was carried out, followed by compression-coded images by compression algorithm based on discrete cosine transform as in standard JPEG with different parameters. This coding was done in different color models: for the experiment were chosen model, RGB and YUV 4: 4: 4.
As for the original form recovery of transformed image blocks necessary to store metadata coding parameters for evaluating the effectiveness of this method is also compressed metadata files long series of algorithms (RLE) and Huffman which are used in the standards: ARJ, ZIP, RAR.
ression results in Kbytes
Source frame RGB YUV 4:4:4
Compression Initial Compressed The amount of the error The amount of the error
size frame (Kb) 100% 80% 20% 1% 100% 80% 20% 1%
Lossless 317 319 322 334 308 330 330 333 315
20 times 1,1 MB 90 90.9 91.8 89.3 74.2 69.3 69.3 67 53.7
50 times 50 50.4 51 47.1 36.6 36.8 36.9 33.4 23.4
Compression metadata files
ARJ 1.71 4.24 57.5 223 1.26 1.42 15.7 126
ZIP 1.86 4.35 57.5 225 1.36 1.51 15.7 128
RAR 1.53 4.25 50.3 170 1.11 1.26 15.5 100
To demonstrate the results obtained when testing the pres- evaluation of its effectiveness, which are shown in Tables 1-2 and ent method, the plot shows the results of compression, as well as histogram (Fig. 4).
Table 2. Evaluation of the plot of compression efficiency
Compression The average amount of personnel in the amount of metadata
RGB YUV 4:4:4
100% 80% 20% 1% 100% 80% 20% 1%
Lossless 320,53 326,25 384,3 478 331,11 331,26 348,5 415
20 times 92,43 96,05 139,6 244,2 70,41 70,56 82,5 153,7
50 times 51,93 55,25 97,4 206,6 37,91 38,16 48,9 123,4
Compression efficiency
RGB YUV 4:4:4
100% 80% 20% 1% 100% 80% 20% 1%
Lossless 0,99 0,97 0,82 0,66 0,96 0,96 0,91 0,76
20 times 0,97 0,94 0,64 0,37 1,28 1,28 1,09 0,59
50 times 0,96 0,9 0,51 0,24 1,32 1,31 1,02 0,41
Figure 4. Assessment efficiency when image compression processing in different color models
This graph shows that the highest efficiency in coding compresses the image in YUV color model 4: 4: 4. This is due to the fact that this model provides a better grip when using the DCP by using different color difference components. As is known, the human eye is more sensitive to color brightness than to its color components. Model 4: 4: 4 is used as a time when imaging, are saturated with a rather small details and sharp transitions in color difference components.
The effectiveness of image reconstruction was estimated standard deviation ofMSE (mean square error) and PSNR (Peak Signal to Noise Ratio), which are calculated according to the formulas [5]:
SNR = 20 ■ log1(
255
4msE'
(2)
TOrgk
MSE =
-Re s,
w-h
(1)
According to the calculations of all the studies, the compressed image reconstruction error with the same error recovery algorithms applied JPEG, i. e. the proposed conversion method does not affect image quality.
Conclusion. The proposed method allows to obtain high homogeneity and a correlation between pixels, thereby increasing the compression ratio in some images, however, additional metadata input for further recovery, increases the volume of the final output file. Compression using standard metodannyh archives increases the efficiency of the method. Therefore, the use of archiving algorithms after the proposed brightness changes will involve an increase in the compression ratio.
References:
1. Jian Song, Zhixing Yang, JunWang. Digital Terrestrial Television Broadcasting: Technology and System. IEEE Press. Canada. - 2015. -456.
2. Под редакцией Соатова Х. С., Гаврилов И. А., Рахимов Т.Г, Пузий А. Н., Носиров Х. Х., Кадиров Ш. М. Цифровое телевидение: Top Image Media press, - Tashkent, - 2016.
3. Носиров Х. Х., Уменьшение размеров изображений для увеличения коэффициента сжатия, и его влияние на качество восстановления в кодеке Дирак//Ежемесячный научный журнал Евразийский союз ученых, - № 9 (30)/ - 2016, - часть 2, -Москва, - 2016.
4. Артюшенко В. М., Шелухин О. И., Афонин М. Ю. Цифровое сжатие видеоинформации и звука: И.: - Москва. - 2003. - С. 430.
5. Salomon David. Data Compression: The Complete Reference (4 ed.). Springer. - 2007. - P. 281. - ISBN 978-1846286025. - Retrieved 26 July - 2012.
DOI: http://dx.doi.org/10.20534/ESR-17-1.2-192-194
Bakiev Masharif Ruzmetovich, Professor in the department «Hydrotechnical construction and engineering structures», Tashkent institute of irrigation and melioration (TIIM), Uzbekistan
E-mail: bakiev1947@rambler.ru Kahhorov Uktam Abdurahimovich, Senior teacher at the department «Hydrotechnical construction and engineering structures» Tashkent institute of irrigation and melioration (TIIM), Uzbekistan.
E-mail: uktam-nig@rambler.ru
Estimation of velocity fields beyond transverse dams at the region of flow potential energy restoration
Abstract: Using main clauses of the theory of turbulent jets, the authors introduce relationships to calculate velocity fields and length of restoration region of the potential energy of flow constrained by bilateral floodplain dams.
Keywords: floodplain, channel, momentum conservation equation, discharge conservation equation, traction forces, velocities in channel, velocities at floodplain, length of potential energy restoration.
Rapid population growth in Uzbekistan, reaching 34,7 mln in 2025, will put on agenda the task of effective use of available land resources, including floodplain lands, and also the task guaranteed water supply into irrigation canals. These tasks can be solved by construction of transverse dams on river floodplains, since erecting transvers dams from local material is much cheaper than bank paving longitudinally.
On the other hand, the regional ecology of rivers in the Aral sea basin is worsening. Construction of regulation structures improves the local ecology: prevents riverside lands from water erosion, allows for regulation of floodplain land use, facilitates the drop of river wa-
ter level for the purpose of improving reclamation state of riverside lands, and, in addition, nowadays the question arose about bringing water to Aral by regulating Amudarya river floodplain at its delta.
Meanwhile, the movement of flood itself in channels with immersed floodplain forms under influence of channel and floodplain complex morphology and roughness, kinematic and dynamic interaction of channel and floodplain flows. At present, there is no methodic for designing bilateral transverse dams at floodplains, therefore conducting high-cost research is needed to justify such projects.
