An aspect of the invention provides a method for determining at least one internal quality attribute of an article (102) of agricultural produce. The method includes receiving a plurality of first spectroscopic values obtained from directing low band light in a first wavelength associated to a low band of wavelengths from at least one low band light source (104) at least partly through the article (102) toward at least one detector (120); receiving a plurality of second spectroscopic values obtained from directing high band light in a second wavelength associated to a high band of wavelengths from at least one high band light source (106) at least partly through the article (102) toward the at least one detector (120); determining at least one measured spatial profile associated to the article, the at least one measured spatial profile comprising at least one of a plurality of ratios of respective first spectroscopic values to respective second spectroscopic values, a plurality of ratios of respective second spectroscopic values to respective first spectroscopic values; and determining the at least one internal quality attribute at least partly from a comparison of the at least one measured spatial profile with at least one reference spatial profile associated to a class of articles of agricultural produce.

The present invention relates to a method for identifying fruit shelf life automatically based on thermal imaging Identifying every day the pattern of change in temperature of fruit's thermal image, we can predict the shelf life of fruit which is how many days the fruit will remain edible in an accurate manner without destructing the fruit. In this system, thermal dataset is created comprising of samples of thermal images (206) of fruit taken on every day after harvesting where fruit may be from cold storage or room temperature using thermal imaging device (204). Transfer learning a deep learning technique is applied on this thermal dataset to compute the threshold weights (210) which are then used to classify or predict the fruit shelf life by comparing these pre-trained weights with the features of fruit's thermal image extracted from convolution and poling layer (212) of deep learning model

A farming machine is configured to identify and treat plants in a field. The farming machine includes one or more light sensors for measuring a characteristic of light. The one or more light sensors are coupled to the farming machine and are directed a substantially upwards orientation away from the plants. A control system adjusts settings of an image acquisition system based on a characteristic of light measured by the one or more light sensors. The image acquisition system captures an image of a plant using one or more image sensors coupled to the farming machine, the one or more image sensors directed in a substantially downwards orientation towards the plants. The control system identifies a plant in the image and actuates a treatment mechanism to treat the identified plant.

Methods and apparatus to generate calibration models in a cloud environment are disclosed. An example apparatus includes training circuitry to generate a calibration model based on a correlation of reference data and spectra, the reference data based on physical samples collected by one or more vehicles, the spectra associated with the physical samples, and distribution circuitry to provide, via a network communication, the calibration model to the one or more vehicles.

Information of plant based on color of a surface of plant is acquired from image data obtained by imaging plant, allowing to acquire information of plant at low cost, compared to chlorophyll meter or spectroscopic analyzer. In crop production like rice plant, fertilization management including fertilizer application management like fertilizer amount determination, or other agricultural works, is supported through a smart phone or the like based on data to be observed, like converted leaf color value is calculated from image data obtained by crop imaging. Camera is connected to smart phone. Converted leaf color value can be obtained from image data obtained by imaging leaf of rice plant by camera. Converted leaf color value is transmitted to management server, for example, amount information of applied fertilizer is required in case where converted leaf color value is less than standard, can be obtained as management information for fertilizer application management.

Provided are methods and/or systems having advantages of cost effective, time saving, and informative user-friendly characteristics to accomplish trait introgression. The methods provided comprise determining presence of omega-3 fatty acids (for example docosahexanoic acid or DHA; docosapentaenoic acid or DPA; Alpha linolenic acid or ALA; and eicosapentaenoic acid or EPA) using Fourier Transformed Infra Red (FTIR) spectrum. The use of FTIR enables analysis of the oil contained in the seeds using a multivariate-based Mid-FTIR model. The methods and/or systems provided advantages of non-destructive analysis to provide information to facilitate trait introgression and other breeding applications.

A method for analysis of algae, comprising: receiving a microscopic image of algae by a cloud server (2501), the microscopic image including a scaling pattern for determining a magnification; determining the magnification by the cloud server based on the scaling pattern (2502); and analyzing the microscopic image by the cloud server based on the magnification to obtain an analysis result (2503).

The present invention relates to the use of a Raman spectral signature of nitrate, as a biomarker for an early, real-time diagnosis of nitrogen status in growing plants in a non-invasive or non-destructive way in order to detect nitrogen deficiency before the onset of any visible symptoms. The early, real-time diagnosis of nitrogen deficiency in plants makes it possible to correct nitrogen deficiency for the avoidance of negative effects on the yield and biomass of growing plants or leafy vegetables.

A kaleidoscopic display assembly, comprising: (a) a stand; (b) a kaleidoscope tube supported by the stand; (c) a display container in alignment with the kaleidoscope tube for containing at least a first botanical specimen viewable in a kaleidoscopic pattern through a viewing opening of the kaleidoscope tube; and (d) an aroma container positioned alongside the tube and enclosing an aroma chamber for containing a second botanical specimen. The aroma container has one or more scent ports extending therethrough for sampling an aroma of the second botanical specimen during viewing of the kaleidoscopic pattern.

Optical system includes a front group, light-shielding member, and rear group that are arranged in this order in direction from object side toward image side. The light-shielding member is provided with opening elongated in first direction. The front group does not image the object at the opening in first section parallel to the first direction and forms intermediate image of the object at the opening in second section perpendicular to the first direction. The rear group has diffractive surface that splits light beam that passes through the opening into light beams at different wavelengths in the second section and focuses the light beams on different locations in the second section. Light beam that is emitted from the front group 11 and that enters the opening is non-parallel light in the first section.