A deep learning algorithm for potato tuber hollow heart classification

Novel deep learning algorithms are proposed for hollow heart detection which is an internal tuber defect. Hollow heart is one of many internal defects that decrease the market value of potatoes in the fresh market and food processing sectors. Susceptibility to internal defects like the hollow heart is influenced by genetic and environmental factors so elimination of defect-prone material in potato breeding programs is important. Current methods of evaluation utilize human scoring which is limiting (only collects binary data) relative to the data collection capacity afforded by computer vision or are based upon x-ray transmission techniques that are both expensive and can be hazardous. Automation of defect classification (e.g. hollow heart) from data sets collected using inexpensive, consumer-grade hardware has the potential to increase throughput and reduce bias in public breeding programs. The proposed algorithms consists of ResNet50 as the backbone of the model followed by a shallow fully connected network (FCN). A simple augmentation technique is performed to increase the number of images in the data set. The performance of the proposed algorithm is validated by investigating metrics such as precision and the area under the curve (AUC).

Cell Wall Polysaccharide Composition of Grafted ‘Liberty’ Watermelon With Reduced Incidence of Hollow Heart Defect

Grafting watermelon scions to interspecific squash hybrids has been found to increase fruit firmness. Triploid (seedless) watermelon are prone to hollow heart (HH), an internal fruit disorder characterized by a crack in the placental tissue expanding to a cavity. Although watermelon with lower tissue firmness tend to have a higher HH incidence, associated differences in cell wall polysaccharide composition are unknown. Grafting “Liberty” watermelon to “Carnivor” (interspecific hybrid rootstock, C. moschata × C. maxima) reduced HH 39% and increased tissue firmness by 3 N. Fruit with and without severe HH from both grafted and non-grafted plants were analyzed to determine differences in cell wall polysaccharides associated with grafting and HH. Alcohol insoluble residues (AIR) were sequentially extracted from placental tissue to yield water soluble (WSF), carbonate soluble (CSF), alkali soluble (ASF), or unextractable (UNX) pectic fractions. The CSF was lower in fruit with HH (24.5%) compared to those without HH (27.1%). AIRs were also reduced, hydrolyzed, and acetylated for GC-MS analysis of monosaccharide composition, and a portion of each AIR was methylated prior to hydrolysis and acetylation to produce partially methylated alditol acetates for polysaccharide linkage assembly. No differences in degree of methylation or galacturonic and glucuronic acid concentrations were found. Glucose and galactose were in highest abundance at 75.9 and 82.4 μg⋅mg–1 AIR, respectively, followed by xylose and arabinose (29.3 and 22.0 μg⋅mg–1). Mannose was higher in fruit with HH (p < 0.05) and xylose was highest in fruit from grafted plants (p < 0.05). Mannose is primarily found in heteromannan and rhamnogalacturonan I side chains, while xylose is found in xylogalacturonan or heteroxylan. In watermelon, 34 carbohydrate linkages were identified with galactose, glucose, and arabinose linkages in highest abundance. This represents the most comprehensive polysaccharide linkage analysis to date for watermelon, including the identification of several new linkages. However, total pectin and cell wall composition data could not explain the increased tissue firmness observed in fruit from grafted plants. Nonetheless, grafting onto the interspecific hybrid rootstock decreased the incidence of HH and can be a useful method for growers using HH susceptible cultivars.

University of Florida Potato Variety Trials Spotlight: Elkton

‘Elkton’ is a white-flesh potato variety suitable for chipping directly from the field (Figures 1 and 2). ‘Elkton’ was selected from the USDA-ARS breeding program in Beltsville, Maryland, by Dr. Haynes in 1997. In 2003, seed of ‘Elkton’ was made available for field evaluation under Florida growing conditions. In 60 trials conducted between 2003 and 2018, ‘Elkton’ yielded 112% in comparison with ‘Atlantic’ (Table 1). In these trials, ‘Elkton’ demonstrated resistance to internal heat necrosis and hollow heart, which are common tuber physiological disorders under high temperature growing conditions.https://edis.ifas.ufl.edu/hs1237 This is a minor revision. Originally published April 2014.

Impact of Grafting on Watermelon Fruit Maturity and Quality

Watermelon (Citrullus lanatus) grafting has emerged as a promising biological management approach aimed at increasing tolerance to abiotic stressors, such as unfavorable environmental conditions. These conditions include environments that are too cold, wet, or dry, have soil nutrient deficiency or toxicity and soil or irrigation water salinity. Studies to date indicate that fruit yield and quality may be positively or negatively affected depending on rootstock-scion combination and growing environment. Growers need information regarding the general effect of rootstocks, as well as specific scion-rootstock interactions on fruit maturity and quality so they can select combinations best suited for their environment. This review summarizes the literature on watermelon grafting with a focus on abiotic stress tolerance and fruit maturity and quality with specific reference to hollow heart and hard seed formation, flesh firmness, total soluble solids, and lycopene content.

Predicting Hollow Heart Incidence in Triploid Watermelon (Citrullus lanatus)

In triploid (seedless) watermelon [Citrullus lanatus var. lanatus (Thunb. Matsum. and Nakai)], hollow heart (HH) is a disorder that is expressed as a crack in the center of the fruit that expands to an open cavity. Although HH incidence and severity is part of a screening process for marketable watermelon fruit during cultigen evaluations, HH incidence is highly variable with growing season, even when the best cultural practices are used. Placental tissue firmness is also measured because firmness is related to the marketability of watermelon and may be related to HH. Genetic and environmental factors can influence watermelon HH development, including plant genetics, pollen amount and viability, pollinator activity, and temperature and rainfall fluctuations. We used seedless watermelon cultigen evaluation data collected over 3 years (2012–14) to determine the relationship between germplasm HH and tissue firmness. Transplanted watermelon representing 30 to 44 cultigens per year were grown at the Central Crops Research Station, Clayton, NC, and interplanted with pollenizers ‘Ace’ and/or ‘SP-6’. Harvested fruit were cut length-wise and rated for HH incidence and severity. Flesh firmness was determined by a handheld penetrometer at five locations in the flesh (stem end, top side, ground spot, blossom end, and heart). A common cultigen subset, consisting of 13 cultigens that were grown in all three experiments, was used for analysis of HH severity and incidence, and placental firmness. The presence of HH was negatively correlated with tissue firmness in both the large multiyear cultigen set (R2 = −0.32; P = 0.0001) and the common cultigen set (R2 = −0.78; P = 0.0001). Cultigens with lower watermelon tissue firmness values had higher HH incidence and severity. By using multiyear cultigen studies and logistic regression, we were able to detect trends for cultigen susceptibility to this highly variable disorder. Using logistic regression, the probability of HH development was highest for ‘Bold Ruler’, ‘Liberty’, and ‘Affirmed’, and lowest for ‘Maxima’ and ‘Captivation’. The identification of cultigens with a tendency for higher or lower rates of HH will be useful for further research of the causes of HH. Measurements of placental flesh firmness may be useful indicators of susceptible cultigens.

Rootstock and Plastic Mulch Effect on Watermelon Flowering and Fruit Maturity in a Verticillium dahliae–Infested Field

Separately, grafting and the use of plastic mulch can increase yield, quality, and early harvest of watermelon (Citrullus lanatus), especially when plants are under biotic and/or abiotic stress. A 2-year field study was conducted to evaluate the combination of four different rootstocks and two types of plastic mulch (black and clear) on date of watermelon first flowering, fruit ripening, yield, and fruit quality when plants were exposed to Verticillium dahliae. Seedless watermelon cv. Secretariat was grafted onto rootstocks Lagenaria siceraria cv. Pelop, Benincasa hispida cv. Round, and two interspecific hybrid squash rootstocks Cucurbita maxima × C. moschata cvs. Super Shintosa and Tetsukabuto, with nongrafted ‘Secretariat’ as the control. Fruit were harvested 0, 7, and 14 days after both the leaflet and tendril attached to the fruit pedicel were completely dry (fruit considered to be physiologically mature). The area under the disease progress curve (AUDPC) values for verticillium wilt were not different for mulch type in either year, although the overall AUDPC value was greatly reduced in the four grafted treatments (227) compared with nongrafted (743). There was no difference in days to male or female flowering due to mulch type or year, and rootstock did not affect first flowering of male flowers. Female flowering was 14 and 11 days later in 2018 and 2019, respectively, for ‘Secretariat’ grafted onto bottle gourd ‘Round’ compared with ‘Secretariat’ grafted onto ‘Tetsukabuto’. Female flowering of ‘Secretariat’ on ‘Round’ was also 7 days later compared with nongrafted ‘Secretariat’ both years. However, days to first harvest was not different with mulch or rootstock and was 92 days after transplanting (DAT) in 2018 and 114 DAT in 2019. There was no difference in yield (fruit number and weight) due to year, harvest date, or mulch, but there was a difference due to grafting. ‘Secretariat’ grafted onto ‘Super Shintosa’ had the greatest total number and weight of fruit per plant (3.7 and 14.8 kg, respectively), and nongrafted ‘Secretariat’ had the lowest (0.7 and 3.2 kg, respectively). Fruit quality attributes hollow heart formation (rating 3.2/5 on average), hard seed count (6 on average), total soluble solids (11% on average), and lycopene content were not different among mulch type, rootstock treatment, or harvest date; however, lycopene content did differ due to year (52.44 and 32.51 µg·g−1 in 2018 and 2019, respectively). Flesh firmness was highest for watermelon grafted onto ‘Super Shintosa’ rootstock (6.7 N) and lowest for nongrafted watermelon (4.3 N). Overall, rootstocks reduced verticillium wilt severity and increased fruit yield whereas mulch had no effects, and 5 V. dahliae colony forming units (cfu)/g of soil may be the minimum level for impact on watermelon fruit yield.