Search

Additional file 2: Supplementary Figure S1. SNP distribution. The number of SNPs within a 1 Mb window size distribute across all chromosomes. The color from green to red represents the gradual increasing SNP number in each window. Supplementary Figure S2. Principal Component Analysis. PCA was conducted on the SNP information of each sample using the GCTA tool. The blue, red, green, cyan and black dots represent SL1 - SL5 populations, respectively. Supplementary Figure S3. Linkage disequilibrium analysis. The numbers in each diamond-shaped box represent the correlation coefficients R2 between two SNPs, and the R2 > 0.2 was considered to indicate linkage. Supplementary Figure S4-S15. GWAS based on body weight from birth (Figure S4) to 110 days old ( Figure S15).

Additional file 2: Figure S1. The morphology of different knob phenotype. (A) S, Lion head goose, (B) W, Sichuan White goose, (C) L, Landes goose. Knob is circled in knob goose, and the position of knob in non-knob goose is indicated using arrow. Figure S2. Histological parameters of the knob skin in three domestic geese breeds. Different letters indicate significant differences among different breeds at P < 0.05. S, Lion head goose; W, Sichuan White goose; L, Landes goose. Figure S3. Principal component analysis (PCA) of the knob skin and bone in three geese breeds, respectively. (A) Skin located on knob. (B) Bone located in knob. SP, skin located on knob in Lion head goose; WP, skin located on knob in Sichuan White goose; LP, skin located on knob in Landes goose; SG, bone located in knob in Lion head goose; WG, bone located in knob in Sichuan White goose; LG, bone located in knob in Landes goose. Figure S4. qRT-PCR validation of expression of DEGs and PSGs. The qRT-PCR data of knob geese (S and W) were normalized by that of non-knob geese (L) and signified by dashed line, while the RNA-seq data were signified by solid line. SP, skin located on knob in Lion head goose; WP, skin located on knob in Sichuan White goose; LP, skin located on knob in Landes goose; SG, bone located in knob in Lion head goose; WG, bone located in knob in Sichuan White goose; LG, bone located in knob in Landes goose. DEGs, differentially expressed genes; PSGs, candidate genes under positive selection.

Due to the complexity of camera viewing angle, background and different execution speed, the difference between hand movements is low, which makes it difficult to extract features in the process of sign language recognition. In order to solve this problem, we propose a sign language recognition framework based on attention mechanism, which enables the network to pay attention to the areas that contain objects related to the activities under consideration. We use Gradient-weighted Class Activation Mapping in the pre-trained CNN for general image recognition to learn the attention map of each frame, and use convolutional LSTM to encode the video in time and space. In view of the fact that the convolution structure in LSTM has no obvious contribution to spatio-temporal feature fusion, the LSTM is adjusted to make it possible to effectively fuse spatiotemporal features with less parameters and computation.

Human robot interaction is playing significant roles in many industries, in which humans are allowed to corporate with robots or control robots to complete some tasks. Achieving realtime control of robots by demonstrating human arm motion is one of the effective methods to improve the quality of human robot interaction. In this paper, a new method to accomplish human skeleton key points extraction is proposed and a system for controlling a virtual UR5 robot (which is modeling in Unity 3D) is developed to imitate human motions. A sequential architecture composed of convolutional networks is used to identify the 2D coordinates; the depth information from Kinect v2 sensor is used to obtain the actual 3D coordinates and the result of the proposed method is used for joint angles calculation. The appropriate mapping relationships between human arm and UR5 robots are built by transforming the arm angel to UR5's degrees of freedom. The performance of this method is measured by the degree of real-time simulation of human hand trajectory. Further, the comparison between the method in this paper and other method is implemented. The experimental results indicate that the developed system and proposed method are reliable and efficient to accomplish human motion imitation.

In this paper we propose a new polarimetric calibration method for C-band spaceborne SAR based on scattering model of amazon rainforest. The well-known Freeman-Durden model is used to express the polarimetric scattering mechanisms of amazon rainforest for C-band spaceborne SAR, and then nonlinear equations with polarimetric distortion parameters are built by use of covariance matrix and solved iteratively to get the polarinetric distortion matrix (PDM). In comparison with the Quegan method, no approximation is used in our method and therefore we can solve all the polarimetric distortion parameters more accurately, even if the crosstalk is high. Moreover, both channel imbalances and crosstalks can be directly estimated by our method, which is more practical as no artificial reflectors are needed to determine the channel imbalances. Simulations based on the RADARSAT-2 data confirm the validity and feasibility of the proposed method.