Yu. V. Kruchinina
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Yu. V. Kruchinina has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
N. P. Goncharov
Institute of Cytology and Genetics of the Siberian Branch of the Russian Academy of Sciences; Novosibirsk State University
2 shared publicationsA. A. Paulish
Novosibirsk State University
1 shared publicationA. Yu. Prikhodko
Novosibirsk State University
1 shared publication- Artem Yu. Pronozin
Institute of Cytology and Genetics of the Siberian Branch of the Russian Academy of Sciences
1 shared publication B. F. Nemtsev
Siberian Research Institute of Plant Production and Breeding – Branch of the Institute of Cytology and Genetics of the Siberian Branch of the Russian Academy of Sciences
1 shared publicationChenxi Zhou
Novosibirsk State University
1 shared publication
Latest Publications
- Open Access
- 5 March 2026
- Published by Institute of Cytology and Genetics, SB RAS in Vavilov Journal of Genetics and Breeding
- Vol. 30 (1) , 27-35
Wheat spikelet detection on RGB images using deep machine learning
This study addresses the challenge of automated high-throughput phenotyping of wheat spike characteristics using modern computer vision and deep learning methods. Accurate estimation of spikelet number is a key indicator of plant productivity, yet traditional manual counting approaches are labor-intensive, slow, and difficult to scale to large breeding datasets. To overcome these limitations, we propose a spikelet detection strategy based on simplified point annotations, where an expert marks only the centers of spikelets rather than drawing detailed segmentation masks or bounding boxes. This significantly reduces annotation time and lowers the overall cost of preparing training datasets for machine learning models. To determine the most effective way of utilizing such simplified annotations, three computational methods were explored: segmentation of binary masks using a U-Net architecture, density regression based on two-dimensional Gaussian distributions optimized via Kullback–Leibler divergence, and detection of fixed-size bounding regions using the YOLOv8 object detection framework. The models were evaluated on dedicated test datasets using both quantitative metrics (MAE, MAPE) and spatial localization metrics (Precision, Recall, F1 score). The results demonstrate that U-Net-based approaches provide consistently high accuracy in spikelet localization and counting while maintaining robustness to annotation imperfections. In contrast, the YOLOv8-based method showed reduced performance, likely due to the geometric mismatch between fixed-size boxes and the natural elongated shape of spikelets. Overall, the proposed methodology highlights the effectiveness of combining minimalistic point-level annotation with advanced segmentation models for automating phenotyping workflows. This approach has the potential to accelerate breeding programs, enhance the efficiency of largescale phenotypic data collection, and support further development of robust computer-vision tools for plant science applications.
- 5 March 2026
- Published by Institute of Cytology and Genetics, SB RAS in Vavilov Journal of Genetics and Breeding
- Vol. 30 (1) , 27-35
Open AccessWheat spikelet detection on RGB images using deep machine learning
This study addresses the challenge of automated high-throughput phenotyping of wheat spike characteristics using modern computer vision and deep learning methods. Accurate estimation of spikelet number is a key indicator of plant productivity, yet traditional manual counting approaches are labor-intensive, slow, and difficult to scale to large breeding datasets. To overcome these limitations, we propose a spikelet detection strategy based on simplified point annotations, where an expert marks only the centers of spikelets rather than drawing detailed segmentation masks or bounding boxes. This significantly reduces annotation time and lowers the overall cost of preparing training datasets for machine learning models. To determine the most effective way of utilizing such simplified annotations, three computational methods were explored: segmentation of binary masks using a U-Net architecture, density regression based on two-dimensional Gaussian distributions optimized via Kullback–Leibler divergence, and detection of fixed-size bounding regions using the YOLOv8 object detection framework. The models were evaluated on dedicated test datasets using both quantitative metrics (MAE, MAPE) and spatial localization metrics (Precision, Recall, F1 score). The results demonstrate that U-Net-based approaches provide consistently high accuracy in spikelet localization and counting while maintaining robustness to annotation imperfections. In contrast, the YOLOv8-based method showed reduced performance, likely due to the geometric mismatch between fixed-size boxes and the natural elongated shape of spikelets. Overall, the proposed methodology highlights the effectiveness of combining minimalistic point-level annotation with advanced segmentation models for automating phenotyping workflows. This approach has the potential to accelerate breeding programs, enhance the efficiency of largescale phenotypic data collection, and support further development of robust computer-vision tools for plant science applications.
- Open Access
- 30 November 2022
- Published by Institute of Cytology and Genetics, SB RAS in Vavilov Journal of Genetics and Breeding
- Vol. 26 (7) , 662-674
Earliness and morphotypes of common wheat cultivars of Western and Eastern Siberia
Скороспелость и морфотип сортов мягкой пшеницы Западной и Восточной Сибири
- 30 November 2022
- Published by Institute of Cytology and Genetics, SB RAS in Vavilov Journal of Genetics and Breeding
- Vol. 26 (7) , 662-674
Open AccessEarliness and morphotypes of common wheat cultivars of Western and Eastern Siberia
Скороспелость и морфотип сортов мягкой пшеницы Западной и Восточной Сибири
- Open Access
- 16 March 2021
- Published by Institute of Cytology and Genetics, SB RAS in Vavilov Journal of Genetics and Breeding
- Vol. 25 (1) , 71-81
Automatic morphology phenotyping of tetra- and hexaploid wheat spike using computer vision methods
Intraspecific classification of cultivated plants is necessary for the conservation of biological diversity, study of their origin and their phylogeny. The modern cultivated wheat species originated from three wild diploid ancestors as a result of several rounds of genome doubling and are represented by di-, tetra- and hexaploid species. The identification of wheat ploidy level is one of the main stages of their taxonomy. Such classification is possible based on visual analysis of the wheat spike traits. The aim of this study is to investigate the morphological characteristics of spikes for hexa- and tetraploid wheat species based on the method of high-performance phenotyping. Phenotyping of the quantitative characteristics of the spike of 17 wheat species (595 plants, 3348 images), including eight tetraploids (Triticum aethiopicum, T. dicoccoides, T. dicoccum, T. durum, T. militinae, T. polonicum, T. timopheevii, and T. turgidum) and nine hexaploids (T. compactum, T. aestivum, i:ANK-23 (near-isogenic line of T. aestivum cv. Novosibirskaya 67), T. antiquorum, T. spelta (including cv. Rother Sommer Kolben), T. petropavlovskyi, T. yunnanense, T. macha, T. sphaerococcum, and T. vavilovii), was performed. Wheat spike morphology was described on the basis of nine quantitative traits including shape, size and awns area of the spike. The traits were obtained as a result of image analysis using the WERecognizer program. A cluster analysis of plants according to the characteristics of the spike shape and comparison of their distributions in tetraploid and hexaploid species showed a higher variability of traits in hexaploid species compared to tetraploid ones. At the same time, the species themselves form two clusters in the visual characteristics of the spike. One type is predominantly hexaploid species (with the exception of one tetraploid, T. dicoccoides). The other group includes tetraploid ones (with the exception of three hexaploid ones, T. compactum, T. antiquorum, T. sphaerococcum, and i:ANK-23). Thus, it has been shown that the morphological characteristics of spikes for hexaploid and tetraploid wheat species, obtained on the basis of computer analysis of images, include differences, which are further used to develop methods for plant classifications by ploidy level and their species in an automatic mode.
- 16 March 2021
- Published by Institute of Cytology and Genetics, SB RAS in Vavilov Journal of Genetics and Breeding
- Vol. 25 (1) , 71-81
Open AccessAutomatic morphology phenotyping of tetra- and hexaploid wheat spike using computer vision methods
Intraspecific classification of cultivated plants is necessary for the conservation of biological diversity, study of their origin and their phylogeny. The modern cultivated wheat species originated from three wild diploid ancestors as a result of several rounds of genome doubling and are represented by di-, tetra- and hexaploid species. The identification of wheat ploidy level is one of the main stages of their taxonomy. Such classification is possible based on visual analysis of the wheat spike traits. The aim of this study is to investigate the morphological characteristics of spikes for hexa- and tetraploid wheat species based on the method of high-performance phenotyping. Phenotyping of the quantitative characteristics of the spike of 17 wheat species (595 plants, 3348 images), including eight tetraploids (Triticum aethiopicum, T. dicoccoides, T. dicoccum, T. durum, T. militinae, T. polonicum, T. timopheevii, and T. turgidum) and nine hexaploids (T. compactum, T. aestivum, i:ANK-23 (near-isogenic line of T. aestivum cv. Novosibirskaya 67), T. antiquorum, T. spelta (including cv. Rother Sommer Kolben), T. petropavlovskyi, T. yunnanense, T. macha, T. sphaerococcum, and T. vavilovii), was performed. Wheat spike morphology was described on the basis of nine quantitative traits including shape, size and awns area of the spike. The traits were obtained as a result of image analysis using the WERecognizer program. A cluster analysis of plants according to the characteristics of the spike shape and comparison of their distributions in tetraploid and hexaploid species showed a higher variability of traits in hexaploid species compared to tetraploid ones. At the same time, the species themselves form two clusters in the visual characteristics of the spike. One type is predominantly hexaploid species (with the exception of one tetraploid, T. dicoccoides). The other group includes tetraploid ones (with the exception of three hexaploid ones, T. compactum, T. antiquorum, T. sphaerococcum, and i:ANK-23). Thus, it has been shown that the morphological characteristics of spikes for hexaploid and tetraploid wheat species, obtained on the basis of computer analysis of images, include differences, which are further used to develop methods for plant classifications by ploidy level and their species in an automatic mode.
- Open Access
- 3 January 2019
- Published by Institute of Cytology and Genetics, SB RAS in Vavilov Journal of Genetics and Breeding
- Vol. 22 (8) , 951-956
Development and investigation of common wheat lines of winter cultivar Bezostaya 1 with combinations of dominant alleles of VRN-1 loci
VRNgenes, determining wheat sensitivity to vernalization, are the main genetic system that defnes the duration of the entire growing period and the durations of the main organogenesis phases. To date, several alleles have been described forVRN-1loci, and allelespecifc primers have been developed that allow rapid identifcation of allelic spectra in common wheat varieties and lines. The unequal influence of different alleles ofVRN-1loci on the duration of the growing period has also been shown; however, there is little information on the effect of the combination of different alleles on heading time. In developing genotypes having different alleles of dominantVRNgenes on the base of the same genetic background, it is necessary to study the genetic effects ofVRNgenes on the duration of the growing season and the individual developmental phases, as well as on productivity. Most varieties presently grown in Russia carry the dominant alleles of twoVRN-1genes:Vrn-A1aandVrn-B1aorVrn-B1c; thus, the task was to create lines combining the dominant alleles ofVrn-A1awithVrn-B1aandVrn-B1cagainst the genetic background of the winter variety Bezostaya 1 (Bez1Vrn-A1a/Vrn-B1aand Bez1Vrn-A1a/Vrn-B1c). Homozygous plants were isolated in the F2 generation by using known allelespecifc primers for theVrn-A1andVrn-B1loci. The durations of the tillering–frst node period, which is the key stage determining growing duration, and the period from shoots to heading were signifcantly reduced in lines with a combination of two dominant alleles ofVRN-1loci compared to isogenic lines of Bezostaya 1 with the dominant allelesVrn-B1aandVrn-B1c. The duration of these developmental phases also decreased in the obtained lines as compared to the isogenic line containing the dominantVrn-A1aallele, but the differences were not signifcant. No substantial differences were found in the duration of other growing phases in lines with two dominant alleles of theVRN-1loci as compared to isogenic lines of Bezostaya 1.
- 3 January 2019
- Published by Institute of Cytology and Genetics, SB RAS in Vavilov Journal of Genetics and Breeding
- Vol. 22 (8) , 951-956
Open AccessDevelopment and investigation of common wheat lines of winter cultivar Bezostaya 1 with combinations of dominant alleles of VRN-1 loci
VRNgenes, determining wheat sensitivity to vernalization, are the main genetic system that defnes the duration of the entire growing period and the durations of the main organogenesis phases. To date, several alleles have been described forVRN-1loci, and allelespecifc primers have been developed that allow rapid identifcation of allelic spectra in common wheat varieties and lines. The unequal influence of different alleles ofVRN-1loci on the duration of the growing period has also been shown; however, there is little information on the effect of the combination of different alleles on heading time. In developing genotypes having different alleles of dominantVRNgenes on the base of the same genetic background, it is necessary to study the genetic effects ofVRNgenes on the duration of the growing season and the individual developmental phases, as well as on productivity. Most varieties presently grown in Russia carry the dominant alleles of twoVRN-1genes:Vrn-A1aandVrn-B1aorVrn-B1c; thus, the task was to create lines combining the dominant alleles ofVrn-A1awithVrn-B1aandVrn-B1cagainst the genetic background of the winter variety Bezostaya 1 (Bez1Vrn-A1a/Vrn-B1aand Bez1Vrn-A1a/Vrn-B1c). Homozygous plants were isolated in the F2 generation by using known allelespecifc primers for theVrn-A1andVrn-B1loci. The durations of the tillering–frst node period, which is the key stage determining growing duration, and the period from shoots to heading were signifcantly reduced in lines with a combination of two dominant alleles ofVRN-1loci compared to isogenic lines of Bezostaya 1 with the dominant allelesVrn-B1aandVrn-B1c. The duration of these developmental phases also decreased in the obtained lines as compared to the isogenic line containing the dominantVrn-A1aallele, but the differences were not signifcant. No substantial differences were found in the duration of other growing phases in lines with two dominant alleles of theVRN-1loci as compared to isogenic lines of Bezostaya 1.