Background:
This paper reviews segmentation techniques for 2D ultrasound fetal images.
Fetal anatomy measurements derived from the segmentation results are used to monitor the
growth of the fetus.
</P><P>
Discussion: The segmentation of fetal ultrasound images is a difficult task due to inherent artifacts
and degradation of image quality with gestational age. There are segmentation techniques for particular
biological structures such as head, stomach, and femur. The whole fetal segmentation algorithms
are only very few.
Conclusion:
This paper presents a review of these segmentation techniques and the metrics used to
evaluate them are summarized.
Segmentation of fetal ultrasound image is an important and necessary task in the automation of fetal biometric measurement. Fetal ultrasound image segmentation is tedious because of the fuzzy nature and textured appearance of fetal structures. Hence, texture based Clustering is proposed for segmenting fetal ultrasound images. Clustering is performed using texture properties of the images which are used for segmenting ultrasound images of fetus. Texture based clustering technique can be used for segmenting all fetal anatomies specifically abdomen, the boundaries of which are very vague and difficult to delineate. Synthetic, simulated ultrasound images and 120 ultrasound fetal images were used for validating the method achieving an accuracy of 90%.
Line jitter due to loss of horizontal synchronization from a noisy video source is a particular video artifact. Line jittering or random horizontal displacements of lines in video images occur when the synchronization signals are corrupted in video storage media, or by electromagnetic interference in wireless video transmission which results in a vertical rolling effect in the digitized video. It is however a curious problem which gives rise to some interesting algorithms. A non-linear algorithm based on detection and estimation technique for restoration of video sequences corrupted by jitter is proposed. The algorithm presented here relies on first applying median filter and then adaptive median filtering on the jittered images. The removal of these artifacts is achieved without destroying important image features like edges and details. The concept is extended for the video sequences. A fast block-matching algorithm based on variable shape search is used for the video sequences and finally temporal filtering is applied to this to obtain the restores video sequence. The proposed algorithm produces better results visually and also video quality indexes, 881M, P8NR are better compared to the different image sequences restoration algorithms.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.