Common Latent Space Exploration for Calibration Transfer across Hyperspectral Imaging-Based Phenotyping Systems

Hyperspectral imaging has increasingly been used in high-throughput plant phenotyping systems. Rapid advancement in the field of phenotyping has resulted in a wide array of hyperspectral imaging systems. However, sharing the plant feature prediction models between different phenotyping facilities becomes challenging due to the differences in imaging environments and imaging sensors. Calibration transfer between imaging facilities is crucially important to cope with such changes. Spectral space adjustment methods including direct standardization (DS), its variants (PDS, DPDS) and spectral scale transformation (SST) require the standard samples to be imaged in different facilities. However, in real-world scenarios, imaging the standard samples is practically unattractive. Therefore, in this study, we presented three methods (TCA, c-PCA, and di-PLSR) to transfer the calibration models without requiring the standard samples. In order to compare the performance of proposed approaches, maize plants were imaged in two greenhouse-based HTPP systems using two pushbroom-style hyperspectral cameras covering the visible near-infrared range. We tested the proposed methods to transfer nitrogen content (N) and relative water content (RWC) calibration models. The results showed that prediction R2 increased by up to 14.50% and 42.20%, while the reduction in RMSEv was up to 74.49% and 76.72% for RWC and N, respectively. The di-PLSR achieved the best results for almost all the datasets included in this study, with TCA being second. The performance of c-PCA was not at par with the di-PLSR and TCA. Our results showed that the di-PLSR helped to recover the performance of RWC, and N models plummeted due to the differences originating from new imaging systems (sensor type, spectrograph, lens system, spatial resolution, spectral resolution, field of view, bit-depth, frame rate, and exposure time) or lighting conditions. The proposed approaches can alleviate the requirement of developing a new calibration model for a new phenotyping facility or to resort to the spectral space adjustment using the standard samples.

PhenoApp: A mobile tool for plant phenotyping to record field and greenhouse observations

With the ongoing cost decrease of genotyping and sequencing technologies, accurate and fast phenotyping remains the bottleneck in the utilizing of plant genetic resources for breeding and breeding research. Although cost-efficient high-throughput phenotyping platforms are emerging for specific traits and/or species, manual phenotyping is still widely used and is a time- and money-consuming step. Approaches that improve data recording, processing or handling are pivotal steps towards the efficient use of genetic resources and are demanded by the research community. Therefore, we developed PhenoApp, an open-source Android app for tablets and smartphones to facilitate the digital recording of phenotypical data in the field and in greenhouses. It is a versatile tool that offers the possibility to fully customize the descriptors/scales for any possible scenario, also in accordance with international information standards such as MIAPPE (Minimum Information About a Plant Phenotyping Experiment) and FAIR (Findable, Accessible, Interoperable, and Reusable) data principles. Furthermore, PhenoApp enables the use of pre-integrated ready-to-use BBCH (Biologische Bundesanstalt für Land- und Forstwirtschaft, Bundessortenamt und CHemische Industrie) scales for apple, cereals, grapevine, maize, potato, rapeseed and rice. Additional BBCH scales can easily be added. The simple and adaptable structure of input and output files enables an easy data handling by either spreadsheet software or even the integration in the workflow of laboratory information management systems (LIMS). PhenoApp is therefore a decisive contribution to increase efficiency of digital data acquisition in genebank management but also contributes to breeding and breeding research by accelerating the labour intensive and time-consuming acquisition of phenotyping data.

CBM: An IoT Enabled LiDAR Sensor for In-Field Crop Height and Biomass Measurements

The phenotypic characterization of crop genotypes is an essential, yet challenging, aspect of crop management and agriculture research. Digital sensing technologies are rapidly advancing plant phenotyping and speeding-up crop breeding outcomes. However, off-the-shelf sensors might not be fully applicable and suitable for agricultural research due to the diversity in crop species and specific needs during plant breeding selections. Customized sensing systems with specialized sensor hardware and software architecture provide a powerful and low-cost solution. This study designed and developed a fully integrated Raspberry Pi-based LiDAR sensor named CropBioMass (CBM), enabled by internet of things to provide a complete end-to-end pipeline. The CBM is a low-cost sensor, provides high-throughput seamless data collection in field, small data footprint, injection of data onto the remote server, and automated data processing. The phenotypic traits of crop fresh biomass, dry biomass, and plant height that were estimated by CBM data had high correlation with ground truth manual measurements in a wheat field trial. The CBM is readily applicable for high-throughput plant phenotyping, crop monitoring, and management for precision agricultural applications.

Corn Leaf Disease Detection with Pertinent Feature Selection Model Using Machine Learning Technique with Efficient Spot Tagging Model

Crop diseases constitute a substantial threat to food safety but, due to the lack of a critical basis, their rapid identification in many parts of the world is challenging. The development of accurate techniques in the field of image categorization based on leaves produced excellent results. Plant phenotyping for plant growth monitoring is an important aspect of plant characterization. Early detection of leaf diseases is crucial for efficient crop output in agriculture. Pests and diseases cause crop harm or destruction of a section of the plant, leading to lower food productivity. In addition, in a number of less-developed countries, awareness of pesticide management and control, as well as diseases, is limited. Some of the main reasons for decreasing food production are toxic diseases, poor disease control and extreme climate changes. The quality of farm crops may be influenced by bacterial spot, late blight, septoria and curved yellow leaf diseases. Because of automatic leaf disease classification systems, action is easy after leaf disease signs are detected. Applying image processing and machine learning methodologies, this research offers an efficient Spot Tagging Leaf Disease Detection with Pertinent Feature Selection Model using Machine Learning Technique (SPLDPFS-MLT). Different diseases deplete chlorophyll in leaves generating dark patches on the surface of the leaf. Machine learning algorithms can be used to identify image pre-processing, image segmentation, feature extraction and classification. Compared with traditional models, the proposed model shows that the model performance is better than those existing.

Functional–Structural Plant Models Mission in Advancing Crop Science: Opportunities and Prospects

Functional–structural plant models (FSPMs) have been evolving for over 2 decades and their future development, to some extent, depends on the value of potential applications in crop science. To date, stabilizing crop production by identifying valuable traits for novel cultivars adapted to adverse environments is topical in crop science. Thus, this study will examine how FSPMs are able to address new challenges in crop science for sustainable crop production. FSPMs developed to simulate organogenesis, morphogenesis, and physiological activities under various environments and are amenable to downscale to the tissue, cellular, and molecular level or upscale to the whole plant and ecological level. In a modeling framework with independent and interactive modules, advanced algorithms provide morphophysiological details at various scales. FSPMs are shown to be able to: (i) provide crop ideotypes efficiently for optimizing the resource distribution and use for greater productivity and less disease risk, (ii) guide molecular design breeding via linking molecular basis to plant phenotypes as well as enrich crop models with an additional architectural dimension to assist breeding, and (iii) interact with plant phenotyping for molecular breeding in embracing three-dimensional (3D) architectural traits. This study illustrates that FSPMs have great prospects in speeding up precision breeding for specific environments due to the capacity for guiding and integrating ideotypes, phenotyping, molecular design, and linking molecular basis to target phenotypes. Consequently, the promising great applications of FSPMs in crop science will, in turn, accelerate their evolution and vice versa.

Multicolor Fluorescence Imaging for the Early Detection of Salt Stress in Arabidopsis

Salt stress is one of the abiotic factors that causes adverse effects in plants and there is an urgent need to detect salt stress in plants as early as possible. Multicolor fluorescence imaging, as a powerful tool in plant phenotyping, can provide information about primary and secondary metabolism in plants to detect the responses of the plants exposed to stress in the early stage. The purpose of this study was to evaluate the potential of multicolor fluorescence imaging’s application in the early detection of salt stress in plants. In this study, the measurements were conducted on Arabidopsis and the multicolor fluorescence images were acquired at 440, 520, 690, and 740 nm with a self-developed imaging system consisting of a UV light-emitting diode (LED) panel for an excitation at 365 nm, a charge coupled device (CCD) camera, interference filters, and a computer. We developed a classification method using the imaging analysis of multicolor fluorescence based on principal component analysis (PCA) and a support vector machine (SVM). The results showed that the four principal fluorescence feature combinations were the ideal indicators as the inputs of the SVM model, and the classification accuracies of the control and salt-stress treatment at 5 days and 9 days were 92.65% and 98.53%, respectively. The results indicated that multicolor fluorescence imaging combined with PCA and SVM could act as a tool for early detection in salt-stressed plants.

OPEN leaf: an open-source cloud-based phenotyping system for tracking dynamic changes at leaf-specific resolution in Arabidopsis

The first draft of the Arabidopsis genome was released more than 20 years ago and despite intensive molecular research, more than 30% of Arabidopsis genes remained uncharacterized or without an assigned function. This is in part due to gene redundancy within gene families or the essential nature of genes, where their deletion results in lethality (i.e., the dark genome). High-throughput plant phenotyping (HTPP) offers an automated and unbiased approach to characterize subtle or transient phenotypes resulting from gene redundancy or inducible gene silencing; however, commercial HTPP platforms remain unaffordable. Here we describe the design and implementation of OPEN leaf, an open-source HTPP system with cloud connectivity and remote bilateral communication to facilitate data collection, sharing and processing. OPEN leaf, coupled with the SMART imaging processing package was able to consistently document and quantify dynamic morphological changes over time at the whole rosette level and also at leaf-specific resolution when plants experienced changes in nutrient availability. The modular design of OPEN leaf allows for additional sensor integration. Notably, our data demonstrate that VIS sensors remain underutilized and can be used in high-throughput screens to identify characterize previously unidentified phenotypes in a leaf-specific manner.