Superpixel as an important pre-processing technique has been successfully used in many vision applications. In this paper, we proposed a region merging method to improve superpixel segmentation accuracy with low compu...Superpixel as an important pre-processing technique has been successfully used in many vision applications. In this paper, we proposed a region merging method to improve superpixel segmentation accuracy with low computational cost. We first segmented the image into many accurate small regions, and then progressively agglomerated them until the desired region number was reached. The region merging weight was derived from a novel energy function, which encourages the superpixel with color consistency and similar size. Experimental results on the Berkeley BSDS500 data set showed that our region merging method can significantly improve the accuracy of superpixel segmentation. Moreover, the region merging method only need 50ms to process a 481x321 image on a single Intel i3 CPU at 2.5 GHz.展开更多
<strong>Purpose</strong><span style="font-family:;" "=""><span style="font-family:Verdana;"><strong>: </strong></span><span style=&q...<strong>Purpose</strong><span style="font-family:;" "=""><span style="font-family:Verdana;"><strong>: </strong></span><span style="font-family:Verdana;">To improve the liver auto-segmentation performance of three-</span><span style="font-family:Verdana;">dimensional (3D) U-net by replacing the conventional up-sampling convolution layers with the Pixel De-convolutional Network (PDN) that considers spatial features. </span><b><span style="font-family:Verdana;">Methods</span></b><span style="font-family:Verdana;">: The U-net was originally developed to segment neuronal structure with outstanding performance but suffered serious artifacts from indirectly unrelated adjacent pixels in its up-sampling layers. The hypothesis of this study was that the segmentation quality of </span></span><span style="font-family:Verdana;">the </span><span style="font-family:Verdana;">liver could be improved with PDN in which the up-sampling layer was replaced by a pixel de-convolution layer (PDL). Seventy</span><span style="font-family:Verdana;">-</span><span style="font-family:;" "=""><span style="font-family:Verdana;">eight plans of abdominal cancer patients were anonymized and exported. Sixty-two were chosen for training two networks: 1) 3D U-Net, and 2) 3D PDN, by minimizing the Dice loss function. The other sixteen plans were used to test the performance. The similarity Dice and Average Hausdorff Distance (AHD) were calculated and compared between these two networks. </span><b><span style="font-family:Verdana;">Results</span></b><span style="font-family:Verdana;">: The computation time for 62 training cases and 200 training epochs was about 30 minutes for both networks. The segmentation performance was evaluated using the remaining 16 cases. For the Dice score, the mean ± standard deviation were 0.857 ± 0.011 and 0.858 ± 0.015 for the PDN and U-Net, respectively. For the AHD, the mean ± standard deviation were 1.575 ± 0.373 and 1.675 ± 0.769, respectively, corresponding to an improvement of 6.0% and 51.5% of mean and standard deviation for the PDN. </span><b><span style="font-family:Verdana;">Conclusion</span></b><span style="font-family:Verdana;">: The PDN has outperformed the U-Net on liver auto-segmentation. The predicted contours of PDN are more conformal and smoother when compared with</span></span><span style="font-family:Verdana;"> the</span><span style="font-family:Verdana;"> U-Net.</span>展开更多
Vehicle license plate (VLP) character segmentation is an important part of the vehicle license plate recognition system (VLPRS).This paper proposes a least square method (LSM) to treat horizontal tilt and vertical til...Vehicle license plate (VLP) character segmentation is an important part of the vehicle license plate recognition system (VLPRS).This paper proposes a least square method (LSM) to treat horizontal tilt and vertical tilt in VLP images.Auxiliary lines are added into the image (or the tilt-corrected image) to make the separated parts of each Chinese character to be an interconnected region.The noise regions will be eliminated after two fusing images are merged according to the minimum principle of gray values. Then,the characters are segmented by projection method (PM) and the final character images are obtained.The experimental results show that this method features fast processing and good performance in segmentation.展开更多
Tracking and segmentation of moving objects are suffering from many problems including those caused by elimination changes, noise and shadows. A modified algorithm for the adaptive background model is proposed by link...Tracking and segmentation of moving objects are suffering from many problems including those caused by elimination changes, noise and shadows. A modified algorithm for the adaptive background model is proposed by linking Gaussian mixture model with the method of principal component analysis PCA. This approach utilizes the advantage of the PCA method in providing the projections that capture the most relevant pixels for segmentation within the background models. We report the update on both the parameters of the modified method and that of the Gaussian mixture model. The obtained results show the relatively outperform of the integrated method.展开更多
文摘Superpixel as an important pre-processing technique has been successfully used in many vision applications. In this paper, we proposed a region merging method to improve superpixel segmentation accuracy with low computational cost. We first segmented the image into many accurate small regions, and then progressively agglomerated them until the desired region number was reached. The region merging weight was derived from a novel energy function, which encourages the superpixel with color consistency and similar size. Experimental results on the Berkeley BSDS500 data set showed that our region merging method can significantly improve the accuracy of superpixel segmentation. Moreover, the region merging method only need 50ms to process a 481x321 image on a single Intel i3 CPU at 2.5 GHz.
文摘<strong>Purpose</strong><span style="font-family:;" "=""><span style="font-family:Verdana;"><strong>: </strong></span><span style="font-family:Verdana;">To improve the liver auto-segmentation performance of three-</span><span style="font-family:Verdana;">dimensional (3D) U-net by replacing the conventional up-sampling convolution layers with the Pixel De-convolutional Network (PDN) that considers spatial features. </span><b><span style="font-family:Verdana;">Methods</span></b><span style="font-family:Verdana;">: The U-net was originally developed to segment neuronal structure with outstanding performance but suffered serious artifacts from indirectly unrelated adjacent pixels in its up-sampling layers. The hypothesis of this study was that the segmentation quality of </span></span><span style="font-family:Verdana;">the </span><span style="font-family:Verdana;">liver could be improved with PDN in which the up-sampling layer was replaced by a pixel de-convolution layer (PDL). Seventy</span><span style="font-family:Verdana;">-</span><span style="font-family:;" "=""><span style="font-family:Verdana;">eight plans of abdominal cancer patients were anonymized and exported. Sixty-two were chosen for training two networks: 1) 3D U-Net, and 2) 3D PDN, by minimizing the Dice loss function. The other sixteen plans were used to test the performance. The similarity Dice and Average Hausdorff Distance (AHD) were calculated and compared between these two networks. </span><b><span style="font-family:Verdana;">Results</span></b><span style="font-family:Verdana;">: The computation time for 62 training cases and 200 training epochs was about 30 minutes for both networks. The segmentation performance was evaluated using the remaining 16 cases. For the Dice score, the mean ± standard deviation were 0.857 ± 0.011 and 0.858 ± 0.015 for the PDN and U-Net, respectively. For the AHD, the mean ± standard deviation were 1.575 ± 0.373 and 1.675 ± 0.769, respectively, corresponding to an improvement of 6.0% and 51.5% of mean and standard deviation for the PDN. </span><b><span style="font-family:Verdana;">Conclusion</span></b><span style="font-family:Verdana;">: The PDN has outperformed the U-Net on liver auto-segmentation. The predicted contours of PDN are more conformal and smoother when compared with</span></span><span style="font-family:Verdana;"> the</span><span style="font-family:Verdana;"> U-Net.</span>
基金Scientific Research Fund of Hunan Province,PRC (No.07JJ6141)Scientific Research Fund of Hunan Provincial Education Department,PRC (No.05C720).
文摘Vehicle license plate (VLP) character segmentation is an important part of the vehicle license plate recognition system (VLPRS).This paper proposes a least square method (LSM) to treat horizontal tilt and vertical tilt in VLP images.Auxiliary lines are added into the image (or the tilt-corrected image) to make the separated parts of each Chinese character to be an interconnected region.The noise regions will be eliminated after two fusing images are merged according to the minimum principle of gray values. Then,the characters are segmented by projection method (PM) and the final character images are obtained.The experimental results show that this method features fast processing and good performance in segmentation.
基金This work was supported in part by the-National Natural Science Foundation of China (61403342, 61273286, U1509207, 61325019, 113 02195), and Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering (2014KLA09).
文摘Tracking and segmentation of moving objects are suffering from many problems including those caused by elimination changes, noise and shadows. A modified algorithm for the adaptive background model is proposed by linking Gaussian mixture model with the method of principal component analysis PCA. This approach utilizes the advantage of the PCA method in providing the projections that capture the most relevant pixels for segmentation within the background models. We report the update on both the parameters of the modified method and that of the Gaussian mixture model. The obtained results show the relatively outperform of the integrated method.