This paper proposes a low complexity algorithm for calculating the soft decision metric of forward error correction codes. In order to reduce the complexity of LLR calculation, neural network training can be used to fit the relationship between input and output during exact LLR calculation. The trained network can skip the complex training process and directly estimate the LLR value. We use random signal-to-noise ratio (SNR) and received signals as the input layer of the neural network, and use the precisely calculated soft decision information as the output layer of the network. After training, the network can accurately estimate the LLR of the received symbols. The performance of 16QAM and 64QAM transmission system using the proposed algorithm and the exact LLR calculation algorithm in the AWGN channel with polar code as the encoding algorithm are presented, respectively. The simulation results show that the proposed algorithm has a similar bit error rate performance to traditional exact calculation algorithms, but with much lower complexity. The proposed algorithm can be applied in various decoding situations that require LLR calculation.
Instance segmentation is an emerging and challenging task. A lot of previous methods have focused on this field and made great progress. However, complex background and mutual occlusion between instances still hinder the development of instance segmentation. In this paper, we aim to solve foreground confusion for instance segmentation. Based on Mask R-CNN, we first add an additional mask branch, which is parallel to the existing mask branch and expected to predict non-target foreground pixels within the predicted bounding box. To differentiate the two mask branches, the original mask branch is called Target Mask Branch while the additional mask branch is called Non-Target Mask Branch. Besides, in order to enhance the feature interaction and differences between the target mask branch and the non-target mask branch, we design a foreground confusion processing module to connect the dual mask branches. In this way, our method can distinguish the target instance and non-target foreground pixels within the corresponding predicted bounding box effectively, and obtain more accurate segmentation masks. Furthermore, a feature distinguish module is proposed to enhance the differences between features from the dual mask branches. We demonstrate our method's competitiveness on MS COCO dataset, which outperforms previous state-of-the-art instance segmentation methods. In addition, we also add our method into other instance segmentation methods, and the experiment results show our method still can obtain stable gains in performance.