A totagram is produced by an iterative spectral phase recovery process resulting in complete information recovery using special masks, without a reference beam. Using these special masking systems reduce computation time, number of masks, and number of iterations. The special masking system is (1) a unity mask together with one or more bipolar binary masks with elements equal to 1 and −1, or (2) a unity mask together with one or more phase masks, or (3) a unity mask together with one pair of masks or more than one pair of masks having binary amplitudes of 0's and 1's, in which the masks in the pair are complementary to each other with respect to amplitude, or (4) one or more pairs of complementary masks with binary amplitudes of 0's and 1's without a unity mask.

An object recognition apparatus includes a first spectrometer configured to obtain a first type of spectrum data from light scattered, emitted, or reflected from an object; a second spectrometer configured to obtain a second type of spectrum data from the light scattered, emitted, or reflected from the object, the second type of spectrum data being different from the first type of spectrum data; an image sensor configured to obtain image data of the object; and a processor configured to identify the object using data obtained from at least two from among the first spectrometer, the second spectrometer, and the image sensor and using at least two pattern recognition algorithms.

A means to automate, using fuzzy logic, the classification of spectral power distributions of optical radiation for lighting systems, and more particularly horticultural lighting systems, is presented. After inputting the spectral power distribution of optical radiation from one or more light sources, radial basis function weights for the spectral power distribution are determined and fuzzified preparatory to fuzzy logic classification. Fuzzy if-then rules are then applied, and an aggregate of the rule votes from the fuzzy if-then rules applied is used to classify the spectral power distribution. The system utilizes a spectral sensor, a fuzzifier module, a fuzzy rule database, fuzzy rule engine, an output fuzzifier module, and a means of displaying the spectral power distribution classification.

A pattern recognition system including an image gathering unit that gathers at least one digital representation of a field, an image analysis unit that pre-processes the at least one digital representation of a field, an annotation unit that provides a visualization of at least one channel for each of the at least one digital representation of the field, where the image analysis unit generates a plurality of image samples from each of the at least one digital representation of the field, and the image analysis unit splits each of the image samples into a plurality of categories.

A spectral camera control device, being installed, along with a spectral camera provided with a liquid crystal tunable filter, in an aircraft capable of stationary flight. The spectral camera control device causes the spectral camera to capture a spectral image in a snapshot mode each time a transmission wavelength of the liquid crystal tunable filter is switched while the aircraft is in stationary flight, and the spectral camera control device causes a plurality of spectral images to be captured in succession at a same transmission wavelength when an SN ratio of the captured spectral image is less than a predetermined threshold.

Provided is an imaging apparatus including: a splitter that splits incident light into pieces of light of two or more wavelength bands; and two or more detectors that detect the pieces of light of two or more wavelength bands respectively and output signals from which wavelengths can be extracted tunably by post-processing.

The invention discloses a hybrid and scanning/snapshot spectral imager operating in both staring-spectral scanning and video snapshot spectral imaging modes. Snapshot spectral imaging operation at a set of selectable critical spectral bands comprise the basis for a machine learning-based estimation and video-rate display of a full hyperspectral cube, without compromising spatial resolution. Operating in the staring-type scanning mode, the disclosed hybrid spectral imager acquires sets of narrow band images at a given tuning step and for a plurality of tuning steps until completing a hyperspectral cube sampling. Scanning operation may be used for optimally configuring snapshot operation, such that redundant information present in the collected spectra is discarded. The disclosed hybrid spectral imager includes embodiments susceptible for miniaturization and low power operation, allowing for their integration into mobile phone and computer platforms. The invention is intended to address applications comprising, at least in part, nondestructive testing, real-time spectral and chemical mapping, noninvasive diagnosis and spectral photography.

The invention provides an automated model training method for training a pipeline for different spectrometers. The automated model training method includes: obtaining first spectral data corresponding to a first spectrometer, and second spectral data corresponding to a second spectrometer; and training the pipeline for the first spectrometer and the second spectrometer according to the first spectral data and the second spectral data, wherein the pipeline corresponds to at least one candidate recognition model. The invention also provides an automated model training device.

In some implementations, a device may determine that an unknown sample is an outlier sample by using an aggregated classification model. The device may determine that one or more spectroscopic measurements are not performed accurately based on determining that the unknown sample is the outlier sample. The device may cause one or more actions based on determining the one or more spectroscopic measurements are not performed accurately.

A spectrometry method used with an apparatus including a spectrometry section, a spectroscopic controller, a spectroscopic image generator, and a display section, the method including generating a teaching-purpose spectroscopic image, generating and displaying a teaching-purpose visualized image, identifying a first teaching area in the teaching-purpose visualized image and generating a first teaching-purpose spectrum, displaying a first icon based on the display color of the first teaching area, accepting teacher data that teaches chromaticity in correspondence with the first icon, generating a conversion rule based on the relationship between the spectrum and the teacher data, generating a measurement-purpose spectrum, and calculating chromaticity based on the conversion rule.