An Accurate Multi-Row Panorama Generation Using Multi-Point Joint Stitching

Zheng, Jin; Zhang, Zhi; Tao, Qiuhao; Shen, Kai; Wang, Yue

doi:10.1109/access.2018.2829082

Cited by 8 publications

(15 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Image stitching is a crucial step in panorama generation. Zheng et al [10] described and explored shooting strategies and image stitching methods in detail. Regarding research on the stitching process, Chen and Tseng [11], by identifying the corresponding feature point in different images (i.e., tie point), determined the tie point quality of panoramic images.…”

Section: Field Of Capturingmentioning

confidence: 99%

“…Kolor Autopano extracts, matches, and transforms feature points into tie points by using the scale-invariant feature transform (SIFT) algorithm [10]. The features extracted using this technique are invariant to image rotation and scaling as well as changes in grayscale values.…”

Section: Homographymentioning

confidence: 99%

“…Therefore, the match with minimum errors must be identified; this process is known as optimization. Kolor Autopano adopts the random sample consensus (RANSAC) algorithm, as shown in Equation ( 5), to minimize the errors between H and all tie points [10]. Specifically, random sample consensus randomly selects tie points as inliers to calculate matrix H and evaluate the errors between the matrix and other tie points.…”

Section: Homographymentioning

confidence: 99%

See 2 more Smart Citations

Panoramic Mapping with Information Technologies for Supporting Engineering Education: A Preliminary Exploration

Lai

Peng

Chang

et al. 2020

IJGI

View full text Add to dashboard Cite

The present researchers took multistation-based panoramic images and imported the processed images into a virtual tour platform to create webpages and a virtual reality environment. The integrated multimedia platform aims to assist students in a surveying practice course. A questionnaire survey was conducted to evaluate the platform’s usefulness to students, and its design was modified according to respondents’ feedback. Panoramic photos were taken using a full-frame digital single-lens reflex camera with an ultra-wide-angle zoom lens mounted on a panoramic instrument. The camera took photos at various angles, generating a visual field with horizontal and vertical viewing angles close to 360°. Multiple overlapping images were stitched to form a complete panoramic image for each capturing station. Image stitching entails extracting feature points to verify the correspondence between the same feature point in different images (i.e., tie points). By calculating the root mean square error of a stitched image, we determined the stitching quality and modified the tie point location when necessary. The root mean square errors of nearly all panoramas were lower than 5 pixels, meeting the recommended stitching standard. Additionally, 92% of the respondents (n = 62) considered the platform helpful for their surveying practice course. We also discussed and provided suggestions for the improvement of panoramic image quality, camera parameter settings, and panoramic image processing.

show abstract

Section: Field Of Capturingmentioning

confidence: 99%

Section: Homographymentioning

confidence: 99%

Section: Homographymentioning

confidence: 99%

See 1 more Smart Citation

Panoramic Mapping with Information Technologies for Supporting Engineering Education: A Preliminary Exploration

Lai

Peng

Chang

et al. 2020

IJGI

View full text Add to dashboard Cite

show abstract

“…Uma das atividades para a construção de um panorama é a costura de imagens, que consiste em unir duas ou mais imagens (NYMAN, 2010). A geração de um panorama simples, com apenas uma faixa de extensão já foi muito estudada e diversos algoritmos que apresentam bons resultados já foram propostos (NYMAN, 2010;SZELISKI, 2006;ZHENG et al, 2018).…”

Section: Introductionunclassified

Construção De Imagens Panorâmicas Em Múltiplas Faixas De Altura E Largura Usando Algoritmos Watershed E Graph-Cut

Silva

Almeida

et al. 2021

COLLOQ EXACTARUM

View full text Add to dashboard Cite

In this work we developed an algorithm for building panoramas with multiple height and width ranges. For multilinear stitching, images were initially placed in a matrix and partial panoramas with images from the same column are generated. To complete the final panorama, the columns were divided with the help of support points and the neighboring columns were stitched, being reassembled at the end of the process. The stitching was performed with a graph cut algorithm in conjunction with the Watershed algorithm.

show abstract

“…Image registration is a process of image matching and transformation of two or more different images. It is widely used in such fields as panoramic image splicing [1,2], high dynamic range imaging [3], simultaneous localization and mapping (SLAM) [4], and so on.…”

Section: Introductionmentioning

confidence: 99%

Image Registration Algorithm Based on Convolutional Neural Network and Local Homography Transformation

Wang

Jiang

et al. 2020

Applied Sciences

View full text Add to dashboard Cite

In order to overcome the poor robustness of traditional image registration algorithms in illuminating and solving the problem of low accuracy of a learning-based image homography matrix estimation algorithm, an image registration algorithm based on convolutional neural network (CNN) and local homography transformation is proposed. Firstly, to ensure the diversity of samples, a sample and label generation method based on moving direct linear transformation (MDLT) is designed. The generated samples and labels can effectively reflect the local characteristics of images and are suitable for training the CNN model with which multiple pairs of local matching points between two images to be registered can be calculated. Then, the local homography matrices between the two images are estimated by using the MDLT and finally the image registration can be realized. The experimental results show that the proposed image registration algorithm achieves higher accuracy than other commonly used algorithms such as the SIFT, ORB, ECC, and APAP algorithms, as well as another two learning-based algorithms, and it has good robustness for different types of illumination imaging.

show abstract

An Accurate Multi-Row Panorama Generation Using Multi-Point Joint Stitching

Cited by 8 publications

References 19 publications

Panoramic Mapping with Information Technologies for Supporting Engineering Education: A Preliminary Exploration

Panoramic Mapping with Information Technologies for Supporting Engineering Education: A Preliminary Exploration

Construção De Imagens Panorâmicas Em Múltiplas Faixas De Altura E Largura Usando Algoritmos Watershed E Graph-Cut

Image Registration Algorithm Based on Convolutional Neural Network and Local Homography Transformation

Contact Info

Product

Resources

About