3results about How to "Improve Grading Accuracy" patented technology

The application relates to a flexible switchable Euphausia superba intelligent sorting method and device, and belongs to the technical field of aquatic product processing and automatic equipment. The application not only realizes high-precision instance segmentation and flexible reference calculation of Euphausia superba through improvement of a deep learning network and a self-adaptive statistical decision model, but also creatively seamlessly converts a dynamic grading signal generated by front-end vision into a physical driving instruction of a servo motor, so that the rear-end spacing of a longitudinal involute sorting pipe is opened and closed in real time and automatically. The application establishes an intelligent linkage closed loop of "visual perception-dynamic decision-physical execution", can automatically and flexibly adjust a physical sorting channel according to real statistical characteristics of batches of materials, and realizes real flexible nondestructive intelligent grading.

The invention discloses an insomnia severity assessment method in combination with a functional near infrared spectrum, and belongs to the technical field of sleep quality. Comprising the following steps: collecting functional near infrared spectrum sequences of a plurality of channels of at least two wavelengths of a prefrontal cortex to obtain three hemoglobin concentration sequences of the plurality of channels; the method comprises the following steps of: mapping a plurality of hemoglobin concentrations into a plurality of predefined brain regions, performing weighted synthesis on channels in the same region to obtain three hemoglobin concentration brain region-level sequences, performing sliding window segmentation to generate a sample set, and calculating a time domain feature vector; obtaining a compact feature vector through feature selection and dimension reduction processing; and constructing a dynamic weight model comprising a plurality of base learners and meta learners, and training the dynamic weight model by using the sample set to obtain a trained dynamic weight model. Compared with equal weight fusion or a single model, the whole grading accuracy and separability are improved by utilizing'brain region entropy weight integration + sliding window feature + dynamic weight Stacking '.