Search

Automatic lane change of vehicles requires driving at the fastest possible speed while ensuring no collision situations. However, regular control is not robust enough to handle unexpected situations or respond to lane separation. To solve these problems, an automatic lane change decision model based on dueling double deep Q-network(D3QN) reinforcement learning model is proposed. The algorithm processes the environmental vehicle information fed back by the internet of vehicles, and then obtains actions through strategies. After the actions are executed, the neural network is trained according to given reward function, and finally the automatic lane change strategy is realized through the trained network and reinforcement learning. The three-lane environment built by Python and the vehicle simulation software CarMaker are used to carry out simulation experiments. The results show that the algorithm proposed has a good control effect, making it feasible and effective. [ABSTRACT FROM AUTHOR]

A conditional generative adversarial network-based high perceptual image compression(HPIC) model was proposed to obtain a compressed image that conforms to human visual perception. In HPIC， the original image was encoded and quantized based on a hyper prior probability model. A generator based on the conditional additional label and residual module was used for the reconstruction of the compressed image， and a discriminator based on the convolutional neural network was used to distinguish the difference between the compressed image and the real image. Based on the bit rate-distortion-perception optimization theory to design the loss function， the perceptual distortion metric was chosen based on the eigenvalues of the pre-trained inception network to achieve image compression reconstruction with high perceptual quality on the one hand， and use the adversarial generation network loss to eliminate compression artifacts and improve the compression accuracy on the other hand. The experimental results show that HPIC achieves a good balance in the bit rate-distortion-perception tradeoff and outperforms the current common algorithms in all perceptual metrics scores， even though the latter uses twice as much bit rate as ours. HPIC can achieve compression with high perceptual quality. [ABSTRACT FROM AUTHOR]

Aiming at multi-focus image fusion, the complementary information at each scale is fused in a targeted manner using a multi-scale fusion method. First, the source image is decomposed into a background layer and a detail layer using the L0 gradient minimization smoothing method, and then a fractional-order edge detection mask is used to preserve the edge information in the background layer, and a fractional-order gradient energy operator is introduced in the detail layer for weight assignment. The experimental results show that the proposed fractional-order gradient energy operator is more sensitive than the integer-order gradient energy clarity operator, which is consistent with the human visual perception. The fusion method proposed effectively avoids artifacts and block effects in the fused images, and retains the source image information more fully. [ABSTRACT FROM AUTHOR]