Various embodiments are disclosed for a machine learning system for automatic genotype selection and performance evaluation using multi-source and spatiotemporal remote sensing data collected from an unmanned aerial system (UAS). A computing device may be configured to access images of a field having a first genotype and a second genotype of at least one crop or plant planted therein. The computing device may apply an image processing routine to the images to analyze the images of the field and determine characteristics of the first genotype and the second genotype of the at least one crop or plant planted in the field. The computing device may then apply a machine learning routine to forecast a first estimated yield of the first genotype and a second estimated yield of the second genotype using the identified characteristics of the first genotype and the second genotype.

A detection device includes a cover, a detection module, and a light source. The cover includes a top wall and a sidewall connected to the top wall. The top wall and the sidewall cooperatively define a receiving cavity. The sidewall includes a first portion close to the top wall and a second portion away from the top wall. The detection module is disposed on the top wall. The light source is disposed on the second portion. The influence of heat on proper operation of the detection device is prevented.

Described herein is a device for monitoring and drying and/or curing process of a coating composition applied at least partially onto at least one surface of a substrate, the device including a movable skid that includes at least one substrate holding unit, at least one lamp, at least one pattern holder including a pattern, and at least one camera, where the substrate holding unit, lamp and pattern holder are positioned on the skid relative to each other such that the light of the lamp can illuminate the pattern of the pattern holder, and the illumination can be projected onto the surface of the substrate, which is held by the substrate holding unit, and where the camera is positioned on the skid relative to the substrate holding unit holding the substrate, such that the camera is capable of recording images of the light reflected off the surface of the substrate.

Disclosed are verification techniques that can be implemented by a device that conducts biological tests to verify that the specimen holding area carries a valid biological specimen. In certain embodiments, the testing device includes a receiving mechanism to receive a carrier, and the carrier includes a holding area that is to carry or to be exposed to a biological specimen. The device can also include a camera module arranged to capture imagery of the carrier, and a processor. In some examples, the processor can capture the imagery of the carrier and identify a visual cue on the carrier. Then, the processor can verify, based on a manner of how the visual cue is displayed in the captured imagery, whether the holding area carries a valid biological specimen.

An image acquisition device including a first plurality of cameras arranged in a mutually spaced configuration, each having a field of view, each field of view lying in a plane and a second plurality of photon emitters arranged in a multiplicity of generally circumferential arrangements about each camera of the first plurality of cameras, at least one photon emitter within the generally circumferential arrangement directing light to a field of view of one of the first plurality of cameras that is not the closest field of view thereto.

Embodiments include a device for testing biological specimen. The device can include a receiving mechanism to receive a carrier. The carrier may include a holding area. The device includes a camera to capture a plurality of images of the holding area. A processor in the device may utilize the camera module to: adaptively select, based on the plurality of images of the holding area, an analytical algorithm suitable for a motion property that is characteristic of the biological specimen being tested; and perform a set of analytic processes that corresponds to the selected analytic algorithm on the captured imagery to generate an analytic result associated with the biological specimen.

The invention relates to a method and device for orienting an umbilicated fruit, in which, during a first orientation phase (22), the presence of at least a portion of an umbilicus is detected in at least one initial image (II), then the fruit is driven (24) in spinning rotation about a first axis of rotation at an angular amplitude of between 5° and 45°, and then the presence of at least a portion of an umbilicus is detected in at least one subsequent image (IU). If at least a portion of an umbilicus is detected in at least one initial image (II) and no longer detected in each subsequent image (IU), the first orientation phase is stopped and the method is continued,

A high-throughput optofluidic device for detecting fluorescent droplets is disclosed. The device uses time-domain encoded optofluidics to detect a high rate of droplets passing through parallel microfluidic channels. A light source modulated with a minimally correlating maximum length sequences is used to illuminate the droplets as they pass through the microfluidic device. By correlating the resulting signal with the expected pattern, each pattern formed by passing droplets can be resolved to identify individual droplets.

A fluid suspended particle classifier system detects particles of one or more classes from a predetermined recognized set of classes based on multiple images of groups of particles. An artificial intelligence classifier combines subtle clues in the images including particle morphology, size, spectral response, fluorescence, movement, density, or aggregation. Class-specific concentration estimations have a low limit of detection, and a high dynamic range. There is a tolerance for mixtures of classes of particles having widely differing sizes and concentrations. Selected particle images, histories and analysis are communicated and stored. Consumable use and manual procedures are minimized or avoided, allowing remote, unattended, real-time operation. In remote water quality monitoring, the appearance of particles resembling the Escherichia coli bacterium, an occasional human pathogen, causes an alarm.

Methods, systems, and apparatus for a camera-enhanced multi-modal gas sensing apparatus including a camera configured to capture imaging data including at least a portion of a test environment including the gas sensing apparatus and an object of interest within the field of view of the camera, multiple gas sensors including a first type of gas sensor and a second type of gas sensor different from the first type of gas sensor which are each sensitive to a respective set of analytes, a housing configured to hold the multiple gas sensors, a gas inlet coupled to the housing and configured to expose the multiple gas sensors to a gas introduced from the test environment via the gas inlet, and a data processing apparatus in data communication with the multiple gas sensors and the camera.