Sowing area under oil-seed rape (Brassica napus L.) has increased during last decade in Latvia and has become more or less stable since 2007. Lack of research on impact of sowing date, sowing rate, fungicide as growth regulator application and other agroecological factor issues on rape autumn growth is observed. The aim of our research, started in autumn 2007 in Research and Study farm Vecauce, was to investigate the influence of applied agroecological factors (five sowing dates, four sowing rates and fungicide (as growth regulator in autumn) application on two type winter rape varieties development in autumn. Winter rape biometrical parameters, dry matter content and dry mass per square m of leaves were estimated. Sowing date affected height of growth point, root neck diameter, plant and root mass, main root length significantly (p is less than 0.05). Sowing rate affected only some indices significantly in autumn 2007. Fungicide (juventus 90 s.c. - metconasol 90 g LE-1) application also affected rape plant biometric parameters during autumn depending on conditions of research year - height of growth point decreased only in year 2007, effect on number of leaves per plant was observed in both trial years, plant weight (2007) and root length (2008) were affected for hybrid variety. Sowing date showed effect on dry matter content in leaves as well as on dry matter mass of leaves per 1 square m. Fungicide application increased dry mass of leaves per 1 square m in two earlier sowing dates. Results are important also for outline of further research directions evaluating significance of above mentioned factors for winter rape wintering in Latvia.

Lack of data about effect of meteorological conditions, sowing date, sowing rate, fungicide as growth regulator application and other agroecological factors on rape autumn growth is observed in Latvia. The aim of our research, started in autumn 2007 and continued up to autumn 2010 in Research and Study farm ''Vecauce'', was to investigate the influence of agroecological factors (sowing date, sowing rate and fungicide (as growth regulator in autumn)) application as well as meteorological conditions on autumn plant growth of two types of winter rape varieties. Winter rape biometrical parameters were estimated, and meteorological parameters (hydrothermal coefficient (HTC) and growing degree days (GDD)) were calculated. Four year results showed that winter oilseed rape seed germination was affected by precipitation and air temperature around the sowing time. Calculated GDD correlated with plant biometrical parameters during trial years. HTC showed moisture effect on rape emergence time. Sowing date was important factor which had strong and significant impact on biometrical parameters of rape plants in autumn. Earlier sowing date increased height of growth point, root neck diameter, plant and root mass, and main root length significantly for both cultivars in four trial years. Such agro-ecological factor as sowing rate (plant density) affected plant biometrical parameters only in some trial years. Plant weight was significantly influenced by sowing rate for both cultivars if sowing rates were equal. Other important parameter - height of growth point - was not significantly influenced by sowing rate for both cultivars when similar sowing rates were used in all years; exception was year 2009 for 'Californium'.

The impact of light penetration into crown and the effect of rootstocks and distance between trees on photosynthetic behaviour were analysed. Apple cultivar ‘Auksis’ was grafted onto dwarfing rootstock P60 and super-dwarfing rootstock P22 and planted at different distances (3 × 1 m, 3 × 0.75 m and 3 × 0.5 m). Nitrogen balance index and photochemical reflectance index were measured at two heights: 0.8 m above ground inside the canopy and 1.5 m above ground outside the canopy; specific leaf area, fresh and dry weight were evaluated from all canopy. The significantly positive effect of dwarfing rootstock P60 on all tested indices was observed comparing to P22 rootstock. Increasing density between trees lead to a decreased specific leaf area and increased nitrogen balance index. The dry and fresh weight ratio and photochemical reflectance index were also significantly affected, but no general tendency was identified. Photochemical reflectance index was not affected by light penetration into canopy, but the nitrogen balance index significantly decreased in the upper part of the canopy. Summarizing, it can be stated that decreasing light penetration into the crown results in an increase in the specific leaf area and photochemical reflectance index, and leads to a decrease in dry and fresh weight ratio and nitrogen balance index.

Deep learning algorithms are becoming default solution for application in business processes where recognition, identification and automated learning are involved. For human identification, analysis of various features can be applied. Face feature analysis is most popular method for identification of person in various stages of life, including children and infants. The aim of this research was to propose deep learning solution for long-term identification of children in educational institutions. Previously proposed conceptual model for long-term re-identification was enhanced. The enhancements include processing of unexpected persons’ scenarios, knowledge base improvements based on results of supervised and unsupervised learning, implementation of video surveillance zones within educational institutions and object tracking results’ data chaining between multiple logical processes. Object tracking results are the solution we found for long-term identification realization.