Reflected light carries ample information about the biochemical composition, tissue architecture, and physiological condition of plants. Recent technical progress has paved the way for affordable imaging hyperspectrometers (IH) providing spatially resolved spectral information on plants on different levels, from individual plant organs to communities. The extraction of sensible information from hyperspectral images is difficult due to inherent complexity of plant tissue and canopy optics, especially when recorded under ambient sunlight. We report on the changes in hyperspectral reflectance accompanying the accumulation of anthocyanins in healthy apple (cultivars Ligol, Gala, Golden Delicious) fruits as well as in fruits affected by pigment breakdown during sunscald development and phytopathogen attacks. The measurements made outdoors with a snapshot IH were compared with traditional “point-type” reflectance measured with a spectrophotometer under controlled illumination conditions. The spectra captured by the IH were suitable for processing using the approaches previously developed for “point-type” apple fruit and leaf reflectance spectra. The validity of this approach was tested by constructing a novel index mBRI (modified browning reflectance index) for detection of tissue damages on the background of the anthocyanin absorption. The index was suggested in the form of mBRI = (R₆₄₀⁻¹ + R₈₀₀⁻¹) − R₆₇₈⁻¹. Difficulties of the interpretation of fruit hyperspectral reflectance images recorded in situ are discussed with possible implications for plant physiology and precision horticulture practices.

Non-invasive techniques for the detection of apple fruit damages are central to the correct operation of sorting lines ensuring storability of the collected fruit batches. The choice of optimal method of fruit imaging and efficient image processing method is still a subject of debate. Here, we have dissected the information content of hyperspectral images focusing on either spectral component, spatial component, or both. We have employed random forest (RF) classifiers using different parameters as inputs: reflectance spectra, vegetation indices (VIs), and spatial texture descriptors (local binary patterns, or LBP), comparing their performance in the task of damage detection in apple fruit. The amount of information in raw hypercubes was found to be over an order of magnitude excessive for the end-to-end problem of classification. Converting spectra to vegetation indices has resulted in a 60-fold compression with no significant loss of information relevant for phenotyping and more robust performance with respect to varying illumination conditions. We concluded that the advanced machine learning approaches could be more efficient if complemented by spectral information about the objects in question. We discuss the potential advantages and pitfalls of the different approaches to the machine learning-based processing of hyperspectral data for fruit grading.

We compared two approaches to non-invasive proximal sensing of the early changes in fresh-cut lettuce leaf quality: hyperspectral imaging and imaging of variable chlorophyll fluorescence contained in the leaves. The estimations made by the imaging techniques were confronted with the quality assessments made by traditional biochemical assays (i.e., relative water content and foliar pigment (chlorophyll and carotenoid) composition. The hyperspectral imaging-based approach provided the highest sensitivity to the decline of fresh-cut lettuce leaf quality taking place within 24 h from cutting. Using of the imaging pulse-amplitude modulated PAM chlorophyll fluorometer was complicated by (i) weak correlation of the spatial distribution pattern of the Qy parameter with the actual physiological condition of the plant object and (ii) its high degree of heterogeneity. Accordingly, the imaging PAM-based approach was sensitive only to the manifestations of leaf quality degradation at advanced stages of the process. Sealing the leaves in polyethylene bags slowed down the leaf quality degradation at the initial stages (<three days) but promoted its rate at more advanced stages, likely due to build-up of ethylene in the bags. An approach was developed to the processing of hyperspectral data for non-invasive monitoring of the lettuce leaves with a potential for implementation in greenhouses and packing lines.

Dormancy is a physiological state that confers winter hardiness to and orchestrates phenological phase progression in temperate perennial plants. Weather fluctuations caused by climate change increasingly disturb dormancy onset and release in plants including tree crops, causing aberrant growth, flowering and fruiting. Research in this field suffers from the lack of affordable non-invasive methods for online dormancy monitoring. We propose an automatic framework for low-cost, long-term, scalable dormancy studies in deciduous plants. It is based on continuous sensing of the photosynthetic activity of shoots via pulse-amplitude-modulated chlorophyll fluorescence sensors connected remotely to a data processing system. The resulting high-resolution time series of JIP-test parameters indicative of the responsiveness of the photosynthetic apparatus to environmental stimuli were subjected to frequency-domain analysis. The proposed approach overcomes the variance coming from diurnal changes of insolation and provides hints on the depth of dormancy. Our approach was validated over three seasons in an apple (Malus × domestica Borkh.) orchard by collating the non-invasive estimations with the results of traditional methods (growing of the cuttings obtained from the trees at different phases of dormancy) and the output of chilling requirement models. We discuss the advantages of the proposed monitoring framework such as prompt detection of frost damage along with its potential limitations.