With the continuous integration of deep learning and the technique of molecular biology, target detection models must accurately detect the position of each cell in the image and classify it correctly. We present a model for the multi‐scale feature fusion of the existing human cell image dataset based on Gaussian mixedly clustering. First, a novel feature extraction network for extracting preliminary features at picture multi scales was presented, which was based on a residual neural network with Instance Normalization and a Mish activation function. Second, the presented model adopts the idea of feature fusion and introduced a new type of feature fusion network to integrate feature graphs on different scales. Furthermore, a Gaussian hybrid clustering algorithm was proposed to cluster the hyperparameters. Based on the experimental results, the average accuracy of the proposed model in the human cell image dataset exceeds 0.96, which improves by 11.9% compared with the existing target detection methods in the same field. Experiments show that the proposed model had been adapted to datasets with uneven sample distribution, providing new ideas for research on medical images.

Background During the COVID-19 pandemic, the accurate forecasting and profiling of the supply of fresh commodities in urban supermarket chains may help the city government make better economic decisions, support activities of daily living, and optimize transportation to support social governance. In urban supermarket chains, the large variety of fresh commodities and the short shelf life of fresh commodities lead to the poor performance of the traditional fresh commodity supply forecasting algorithm. Methods Unlike the classic method of forecasting a single type of fresh commodity, we proposed a third-order exponential regression algorithm incorporating the block Hankle tensor. First, a multi-way delay embedding transform was used to fuse multiple fresh commodities sales to a Hankle tensor, for aggregating the correlation and mutual information of the whole category of fresh commodities. Second, high-order orthogonal iterations were performed for tensor decomposition, which effectively extracted the high-dimensional features of multiple related fresh commodities sales time series. Finally, a tensor quantization third-order exponential regression algorithm was employed to simultaneously predict the sales of multiple correlated fresh produce items. Results The experiment result showed that the provided tensor quantization exponential regression method reduced the normalized root mean square error by 24% and the symmetric mean absolute percentage error by 22%, compared with the state-of-the-art approaches.