Abstract: Methods and systems are described for image segmentation. A machine learning model is applied to a set of images to generate results. The results may be obtained as a probability map for each image in the set of images. The model may be trained by accessing a set of labeled images, each image associated with a label indicating a location of a feature within a respective image. An initial set of parameters is accessed. An encoder is initialized with the initial set of parameters. The encoder is applied to the set of labeled images to generate a prediction of a feature location within each image. The initial set of parameters are updated based on the predictions and the label associated with the labeled images. The updated set of parameters and an additional set of parameters generated using a set of unlabeled images are aggregated.

Abstract: Reactive foils and their uses are provided as localized heat sources useful, for example, in ignition, joining and propulsion. An improved reactive foil is preferably a freestanding multilayered foil structure made up of alternating layers selected from materials that will react with one another in an exothermic and self-propagating reaction. Upon reacting, this foil supplies highly localized heat energy that may be applied, for example, to joining layers, or directly to bulk materials that are to be joined. This foil heat-source allows rapid bonding to occur at room temperature in virtually any environment (e.g., air, vacuum, water, etc.). If a joining material is used, the foil reaction will supply enough heat to melt or soften the joining material, which upon cooling will form a strong bond, joining two or more bulk materials.