Search

Flower recognition has important application value in the field of flower cultivation and planting. As a kind of fine-grained image recognition, the traditional flower recognition has the problem of low recognition accuracy. In order to solve these problems, this paper proposes a neural network algorithm of ResNet structure that integrates CBAM mechanism, and adds residual blocks of attention modules to the second layer to the fifth layer of the ResNet structure. Finally, the results are output through adaptive average pooling and full connection layer. Experimental results show that the recognition accuracy of the model is close to 98% even when the recognition is difficult and there are few training data. Compared with the traditional deep learning model, the model proposed in this paper significantly improves the recognition accuracy.

This paper presents a microfluidic system enabling continuous characterization of specific membrane capacitance (Cspecific membrane) and cytoplasm conductivity (σcytoplasm) of single cells by aspirating cells continuously through a constriction channel with cell elongations and impedance profiles at two frequencies (1 kHz and 100 kHz) measured simultaneously. 1 kHz impedance data were used to evaluate cellular sealing properties and 100 kHz impedance data were used to quantify membrane capacitance and cytoplasm conductance, which were than translated to Cspecific membrane and σcytoplasm based on the proposed equivalent circuit model. Four cell lines were used to evaluate this technique, producing Cspecific membrane and σcytoplasm of 3.81±0.90 μF/cm2 and 0.49±0.07 S/m (CRL-5803, n=489), 4.28±1.05 μF/cm2 and 0.45±0.08 S/m (CCL-185, n=487), 3.67±1.00 μF/cm2 and 0.47±0.09 S/m (CRL-1932, n=302), as well as 4.53±1.51 μF/cm2 and 0.55±0.14 S/m (CRL-1999, n=216), respectively.

This paper presents a microfluidic system enabling continuous characterization of specific membrane capacitance (Cspecific membrane) and cytoplasm conductivity (σcytoplasm) of single biological cells. In this study, cells were aspirated through a constriction channel while cell elongations and impedance profiles at 1 kHz and 100 kHz were measured simultaneously using microscopy imaging and a lock-in amplifier. Based on the proposed equivalent circuit model, raw data were translated to Cspecific membrane and σcytoplasm, which were 3.67±1.00 vs. 4.53±1.51 μF/cm2 and 0.47±0.09 vs. 0.55±0.14 S/m for the kidney tumor cell line of 786-O (n=302) and the vascular smooth muscle cell line of T2 (n=216), respectively.