Model-assisted phenotyping by digital images in papaya breeding program

Manual phenotyping for papaya Carica papaya (L) breeding purposes limits the evaluation of a great number of plants and hampers selection of superior genotypes. This study aimed to validate two methodologies for the phenotyping of morpho-agronomic plant traits using image analysis and fruit traits through image processing. In plants of the THB variety and UENF/Caliman-01 hybrid two images (A and B) were analyzed to estimate commercial and irregularly shaped fruits. Image A was also used in the estimation of plant height, stem diameter and the first fruit insertion height. In THB fruits, largest and smallest diameters, length, and volume were estimated by using a caliper and image processing (IP). Volume was obtained by water column displacement (WCD) and by the expression of ellipsoid approximation (EA). Correlations above 0.85 between manual and image measurements were obtained for all traits. The averages of the morpho-agronomic traits, estimated by using images, were similar when compared to the averages measured manually. In addition, the errors of the proposed methodologies were low compared to manual phenotyping. Bland-Altman's approach indicated agreement between the volume estimated by WCD and EA using caliper and IP. The strong association obtained between volume and fruit weight suggests the use of regression to estimate this trait. Thus, the expectation is that image-based phenotyping can be used to expand the experiments, thereby maintaining accuracy and providing greater genetic gains in the selection of superior genotypes.

Model-assisted phenotyping by digital images in papaya breeding program

Manual phenotyping for papaya Carica papaya (L) breeding purposes limits the evaluation of a great number of plants and hampers selection of superior genotypes. This study aimed to validate two methodologies for the phenotyping of morpho-agronomic plant traits using image analysis and fruit traits through image processing. In plants of the THB variety and UENF/Caliman-01 hybrid two images (A and B) were analyzed to estimate commercial and irregularly shaped fruits. Image A was also used in the estimation of plant height, stem diameter and the first fruit insertion height. In THB fruits, largest and smallest diameters, length, and volume were estimated by using a caliper and image processing (IP). Volume was obtained by water column displacement (WCD) and by the expression of ellipsoid approximation (EA). Correlations above 0.85 between manual and image measurements were obtained for all traits. The averages of the morpho-agronomic traits, estimated by using images, were similar when compared to the averages measured manually. In addition, the errors of the proposed methodologies were low compared to manual phenotyping. Bland-Altman's approach indicated agreement between the volume estimated by WCD and EA using caliper and IP. The strong association obtained between volume and fruit weight suggests the use of regression to estimate this trait. Thus, the expectation is that image-based phenotyping can be used to expand the experiments, thereby maintaining accuracy and providing greater genetic gains in the selection of superior genotypes.(AU)

PROCESSAMENTO DIGITAL DE IMAGENS PARA ANÁLISE DO COMPORTAMENTO E BEM-ESTAR DE FRANGOS DE CORTE

A presente pesquisa foi conduzida com o objetivo principal de utilizar as técnicas de visão computacional para desenvolver um programa que permitisse identificar, qualificar e quantificar a frequência de comportamento ingestivo expressos por frangos de corte. E analisar as diferenças entre o comportamento ingestivo frente aos fatores idade, ambiente de produção, sexagem e sistema de iluminação, utilizando ferramentas estocásticas de análises multivariadas. A validação do software foi realizada pelo resultado obtido a partir da análise visual de um especialista, por meio de análise de regressão linear pela plataforma R. Com base nas análises verificou-se que o coeficiente de determinação (R²) variou de 0,74 a 0,97, para todos os eventos de validação do software, o que evidencia boa caracterização do comportamento ingestivo de frangos de corte pela visão computacional. O comportamento alimentar dos frangos de corte foi influenciado pelo ambiente, principalmente no período da manhã e tarde. A noite foi mais evidente a influência do sistema de iluminação a base de LED branco, pelo maior ganho de peso apresentado pelas aves. As aves submetidas aos sistemas de iluminação não apresentaram alteração comportamental durante o dia, pelo fato da incidência da radiação solar ser maior que a illuminância promovida pelo sistema de iluminação. A variação do ganho de peso e peso vivo das aves apresentaram melhor resposta quando as aves foram expostas a temperatura de 27oC, ou seja, 3°C acima da condição de conforto recomenda para 5ª semana do ciclo de criação.

The present research was conducted with the main objective of using computational vision techniques to develop a program that would identify, qualify and quantify the frequency of ingestive behaviors expressed by broilers. And analyze the differences between ingestive behaviors, observed, facing the factors age, production environment, sexing and lighting system, using stochastic tools of multivariate analysis. The validation of the software was performed by the result obtained from the visual analysis of a specialist, for that was used linear regression analysis through the R platform. Based on the correlation analyzes, it was verified that the coefficient of determination (R²) varied from 0.74 to 0.97, for all software validation events. The number of chickens in the regions of interest presented R² above 0.70 for females and above 0.89 for males, which allowed the characterization of the ingestive behavior of broilers by the computational view. Regarding ingestive behavior as a function of the environmental variables, it can be verified that from 9:00 a.m. to 3:00 p.m. there was a significant increase of all the variables expressed by birds submitted to red and white LED lighting, characterizing an indicative of thermal discomfort, where the 5th week was the hottest week, presenting ITGU values higher than 79 for red LED, reaching 80. And for white LED the values were higher than 77, reaching 81, but the birds suffered with the discomfort at that time during the 3 weeks of experiment. The behavior "drinking male (T1B and T2B)" had a positive association with the thermal characteristics of the environment in the three periods, ie, drinking more water would be a way to alleviate this thermal discomfort. Another fact observed was the variation in the body weight of broiler chickens as a function of the temperature of the environment for the last three weeks of breeding and it was observed that the temperature influenced the quadratic weight gain, where the largest body weights (1864 g in the box with red LED and 1898 g in the box with white LED), were observed in the temperature of 27.14ºC, in the 5th week, verifying an increase of 3.14ºC in the ambient temperature relation with the ambient temperature recommended in the literature as indicative of thermal comfort (1824ºC for 5º week), which concludes that broilers kept at this temperature have a higher feed efficiency.