Image/video processing for fruit detection in the tree using hard-coded feature extraction algorithms has shown high accuracy on fruit detection during recent years. While accurate, these approaches even with high-end hardware are still computationally intensive and too slow for real-time systems. This paper details the use of deep convolution neural networks architecture based on single-stage detectors. Using deep-learning techniques eliminates the need for hard-code specific features for specific fruit shapes, color and/or other attributes. This architecture takes the input image and divides into AxA grid, where A is a configurable hyper-parameter that defines the fineness of the grid. To each grid cell an image detection and localization algorithm is applied. Each of those cells is responsible to predict bounding boxes and confidence score for fruit (apple and pear in the case of this study) detected in that cell. We want this confidence score to be high if a fruit exists in a cell, otherwise to be zero, if no fruit is in the cell. More than 100 images of apple and pear trees were taken. Each tree image with approximately 50 fruits, that at the end resulted on more than 5000 images of apple and pear fruits each. Labeling images for training consisted on manually specifying the bounding boxes for fruits, where (x, y) are the center coordinates of the box and (w, h) are width and height. This architecture showed an accuracy of more than 90% fruit detection. Based on correlation between number of visible fruits, detected fruits on one frame and the real number of fruits on one tree, a model was created to accommodate this error rate. Processing speed is higher than 20 FPS which is fast enough for any grasping/harvesting robotic arm or other real-time applications. HIGHLIGHTS Using new convolutional deep learning techniques based on single-shot detectors to detect and count fruits (apple and pear) within the tree canopy.

Despite the availability of whole genome sequences of apple and peach, there has been a considerable gap between genomics and breeding. To bridge the gap, the European Union funded the FruitBreedomics project (March 2011 to August 2015) involving 28 research institutes and private companies. Three complementary approaches were pursued: (i) tool and software development, (ii) deciphering genetic control of main horticultural traits taking into account allelic diversity and (iii) developing plant materials, tools and methodologies for breeders. Decisive breakthroughs were made including the making available of ready-to-go DNA diagnostic tests for Marker Assisted Breeding, development of new, dense SNP arrays in apple and peach, new phenotypic methods for some complex traits, software for gene/QTL discovery on breeding germplasm via Pedigree Based Analysis (PBA). This resulted in the discovery of highly predictive molecular markers for traits of horticultural interest via PBA and via Genome Wide Association Studies (GWAS) on several European genebank collections. FruitBreedomics also developed pre-breeding plant materials in which multiple sources of resistance were pyramided and software that can support breeders in their selection activities. Through FruitBreedomics, significant progresses were made in the field of apple and peach breeding, genetics, genomics and bioinformatics of which advantage will be made by breeders, germplasm curators and scientists. A major part of the data collected during the project has been stored in the FruitBreedomics database and has been made available to the public. This review covers the scientific discoveries made in this major endeavour, and perspective in the apple and peach breeding and genomics in Europe and beyond.

The expolinear growth model of Goudriaan and Monteith (1990) is proposed as a new model for the inherent growth pattern of fruit of apple (Malus domestica Borkh.), defined as growth pattern under apparently non–limiting conditions This function has three parameters: maximum relative growth rate, maximum absolute growth rate, and “lost time” (x intercept of the linear growth phase). Apple fruit growth (weight basis) at very low crop loads and apparently optimum environmental conditions, displays an early positive curvilinear growth followed by linear growth to harvest, and is described well by the expolinear function. The model also fits growth patterns of ‘Empire’ and ‘Golden Delicious’ apple fruit differing in the rate of growth in the exponential phase due to differences in the crop load. Estimates of cortical cell numbers in ‘Empire’ fruit from related studies suggest that during the linear phase in mid–season, different growth rates among crop load treatments were apparently controlled by differences in numbers of cells in the fruit, since estimated growth rates per cortical cell were essentially constant over several treatments.

The relative contributions of xylem, phloem, and transpiration to fruit growth and the daily patterns of their flows have been determined in peach, during the two stages of rapid diameter increase, by precise and continuous monitoring of fruit diameter variations. Xylem, phloem, and transpiration contributions to growth were quantified by comparing the diurnal patterns of diameter change of fruits, which were then girdled and subsequently detached. Xylem supports peach growth by 70%, and phloem 30%, while transpiration accounts for ∼60% of daily total inflows. These figures and their diurnal patterns were comparable among years, stages, and cultivars. Xylem was functional at both stage I and III, while fruit transpiration was high and strictly dependent on environmental conditions, causing periods of fruit shrinkage. Phloem imports were correlated to fruit shrinkage and appear to facilitate subsequent fruit enlargement. Peach displays a growth mechanism which can be explained on the basis of passive unloading of photoassimilates from the phloem. A pivotal role is played by the large amount of water flowing from the tree to the fruit and from the fruit to the atmosphere.

Light quality (sunlight spectrum) management promises to provide a new technological alternative to sustainable production in horticultural crops. However, little information exists about physiological and technological aspects on light quality management in fruit crops. Sunlight composition changes widely in orchard canopies, inducing different plant responses in fruit trees mediated by phytochrome (PHY) and cryptochrome (CRY) activity. High proportion of far-red (FR) in relation to red (R) light increases shoot elongation, while blue (B) light induces shoot dwarfing. Red and ultraviolet (UV) light increases fruit skin anthocyanin synthesis, while FR light shows a negative effect. Red and B light can also alter leaf morpho-physiological traits in fruit trees, such palisade thickness, stomatal aperture, and chlorophyll content. Besides improvement of photosynthetically active radiation (PAR) availability, the use of reflective films improves UV and R light proportion, with positive effects on PHY mediated-responses (fruit color, fruit weight, shoot growth), as reported in apple (Malus domestica Borkh.), peach (Prunus persica (L.) Batsch), and sweet cherry (Prunus avium (L.) L.). Colored nets widely alter spectral light composition with effects on plant growth, yield, and quality in apple, kiwifruit (Actinidia deliciosa A. Chev. C.F. Liang & A.R. Ferguson), peach, and blueberry (Vaccinium corymbosum L.) orchards. Mechanisms of colored nets seem to be associated to photosynthetic and morphogenetic process regulated by PAR availability, R/B light proportion, and CRY activity. Alteration of light quality affects significantly fruit tree plant responses and could be a useful tool for sustainable (e.g. lower use of chemicals and labor-practices) management of yield and quality in modern orchards.