Extracting the cell objects of red tide algae is the most important step in the construction of an automatic microscopic image recognition system for harmful algal blooms.This paper describes a set of composite method...Extracting the cell objects of red tide algae is the most important step in the construction of an automatic microscopic image recognition system for harmful algal blooms.This paper describes a set of composite methods for the automatic segmentation of cells of red tide algae from microscopic images.Depending on the existence of setae,we classify the common marine red tide algae into non-setae algae species and Chaetoceros,and design segmentation strategies for these two categories according to their morphological characteristics.In view of the varied forms and fuzzy edges of non-setae algae,we propose a new multi-scale detection algorithm for algal cell regions based on border-correlation,and further combine this with morphological operations and an improved GrabCut algorithm to segment single-cell and multicell objects.In this process,similarity detection is introduced to eliminate the pseudo cellular regions.For Chaetoceros,owing to the weak grayscale information of their setae and the low contrast between the setae and background,we propose a cell extraction method based on a gray surface orientation angle model.This method constructs a gray surface vector model,and executes the gray mapping of the orientation angles.The obtained gray values are then reconstructed and linearly stretched.Finally,appropriate morphological processing is conducted to preserve the orientation information and tiny features of the setae.Experimental results demonstrate that the proposed methods can effectively remove noise and accurately extract both categories of algae cell objects possessing a complete shape,regular contour,and clear edge.Compared with other advanced segmentation techniques,our methods are more robust when considering images with different appearances and achieve more satisfactory segmentation effects.展开更多
Segmentation and edge regulation are studied deeply to extract buildings fromDSM data produced in this paper. Building segmentation is the first step to extract buildings, anda new segmentation method-adaptive iterati...Segmentation and edge regulation are studied deeply to extract buildings fromDSM data produced in this paper. Building segmentation is the first step to extract buildings, anda new segmentation method-adaptive iterative segmentation considering rati-o mean square-is proposedto extract the contour of buildings effectively. A sub-image (such as 50X50 pixels) of the image isprocessed in sequence, the average gray level and its ratio mean square are calculated first, thenthreshold of the sub-image is selected by using iterative threshold segmentation. The current pixelis segmented according to the threshold, the average gray level and the ratio mean square of thesub-image. The edge points of the building are grouped according to the azimuth of neighbor points,and then the optimal azimuth of the points that belong to the same group can be calculated by usingline interpolation.展开更多
Extracting building contours from aerial images is a fundamental task in remote sensing.Current building extraction methods cannot accurately extract building contour information and have errors in extracting small-sc...Extracting building contours from aerial images is a fundamental task in remote sensing.Current building extraction methods cannot accurately extract building contour information and have errors in extracting small-scale buildings.This paper introduces a novel dense feature iterative(DFI)fusion network,denoted as DFINet,for extracting building contours.The network uses a DFI decoder to fuse semantic information at different scales and learns the building contour knowledge,producing the last features through iterative fusion.The dense feature fusion(DFF)module combines features at multiple scales.We employ the contour reconstruction(CR)module to access the final predictions.Extensive experiments validate the effectiveness of the DFINet on two different remote sensing datasets,INRIA aerial image dataset and Wuhan University(WHU)building dataset.On the INRIA aerial image dataset,our method achieves the highest intersection over union(IoU),overall accuracy(OA)and F 1 scores compared to other state-of-the-art methods.展开更多
基金Supported by the National Natural Science Foundation of China(Nos.61301240,61271406)
文摘Extracting the cell objects of red tide algae is the most important step in the construction of an automatic microscopic image recognition system for harmful algal blooms.This paper describes a set of composite methods for the automatic segmentation of cells of red tide algae from microscopic images.Depending on the existence of setae,we classify the common marine red tide algae into non-setae algae species and Chaetoceros,and design segmentation strategies for these two categories according to their morphological characteristics.In view of the varied forms and fuzzy edges of non-setae algae,we propose a new multi-scale detection algorithm for algal cell regions based on border-correlation,and further combine this with morphological operations and an improved GrabCut algorithm to segment single-cell and multicell objects.In this process,similarity detection is introduced to eliminate the pseudo cellular regions.For Chaetoceros,owing to the weak grayscale information of their setae and the low contrast between the setae and background,we propose a cell extraction method based on a gray surface orientation angle model.This method constructs a gray surface vector model,and executes the gray mapping of the orientation angles.The obtained gray values are then reconstructed and linearly stretched.Finally,appropriate morphological processing is conducted to preserve the orientation information and tiny features of the setae.Experimental results demonstrate that the proposed methods can effectively remove noise and accurately extract both categories of algae cell objects possessing a complete shape,regular contour,and clear edge.Compared with other advanced segmentation techniques,our methods are more robust when considering images with different appearances and achieve more satisfactory segmentation effects.
基金theNationalNaturalScienceFoundationofChina (No 40 2 0 1 0 35)
文摘Segmentation and edge regulation are studied deeply to extract buildings fromDSM data produced in this paper. Building segmentation is the first step to extract buildings, anda new segmentation method-adaptive iterative segmentation considering rati-o mean square-is proposedto extract the contour of buildings effectively. A sub-image (such as 50X50 pixels) of the image isprocessed in sequence, the average gray level and its ratio mean square are calculated first, thenthreshold of the sub-image is selected by using iterative threshold segmentation. The current pixelis segmented according to the threshold, the average gray level and the ratio mean square of thesub-image. The edge points of the building are grouped according to the azimuth of neighbor points,and then the optimal azimuth of the points that belong to the same group can be calculated by usingline interpolation.
基金National Natural Science Foundation of China(No.61903078)Fundamental Research Funds for the Central Universities,China(No.2232021A-10)+1 种基金Shanghai Sailing Program,China(No.22YF1401300)Natural Science Foundation of Shanghai,China(No.20ZR1400400)。
文摘Extracting building contours from aerial images is a fundamental task in remote sensing.Current building extraction methods cannot accurately extract building contour information and have errors in extracting small-scale buildings.This paper introduces a novel dense feature iterative(DFI)fusion network,denoted as DFINet,for extracting building contours.The network uses a DFI decoder to fuse semantic information at different scales and learns the building contour knowledge,producing the last features through iterative fusion.The dense feature fusion(DFF)module combines features at multiple scales.We employ the contour reconstruction(CR)module to access the final predictions.Extensive experiments validate the effectiveness of the DFINet on two different remote sensing datasets,INRIA aerial image dataset and Wuhan University(WHU)building dataset.On the INRIA aerial image dataset,our method achieves the highest intersection over union(IoU),overall accuracy(OA)and F 1 scores compared to other state-of-the-art methods.