Image mosaicking using SURF features of line segments

Yang, Zhibin; Shen, Dinggang; Yap, Pew Thian

doi:10.1371/journal.pone.0173627

Cited by 20 publications

(12 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2) SURF descriptor The SURF descriptor is obtained by calculating the main direction and feature vector [21] . x y x y…”

Section: Image Mosaic Methodologies 221 the Phase Correlation Algormentioning

confidence: 99%

A fast nursery plug tray images sequence mosaic technique

Tan

et al. 2018

International Journal of Agricultural and Biological Engineering

View full text Add to dashboard Cite

A machine vision based system to estimate the nursery plug tray sowing quantity is necessary in the rice seedling nursery. Because the super hybrid rice is of small diameter, in order to obtain better estimation results of sowing quantity, only part of the tray scene is captured by machine vision system. Therefore, nursery tray image sequence mosaic algorithm is required so as to obtain the whole nursery tray sowing performance. In this paper, a fast nursery plug tray image mosaic technique based on Phase Correlation and Speeded up Robust Features (the PC-SURF) algorithm was introduced. To reduce huge computational complexity in feature extraction and avoid wrong match points from non-overlapping regions between two adjacent images, firstly the proposed method used Phase Correlation to approximately locate the overlapping regions between two sequential images so as to narrow down the processing area for further feature extraction. Then, the SURF algorithm was implemented in the overlapping region of the images to perform image registration and image blending. Image registration and blending were then performed using the RANSAC algorithm and transition smoothing method. Finally, sequential images were warped into a single frame to produce a panoramic tray image with high resolution and better quality. The test results showed that the PC-SURF algorithm greatly outperforms the SURF algorithm in point matching accuracy, time consumption, and image mosaic accuracy. The average feature point matching accuracy of the PC-SURF algorithm was improved by approximately 7.14%. The implementation time was almost three times faster. The Root Mean Squared Error (RMSE) of image blending in the R, G, B channels (RMSEr, RMSEg, and RMSEb) decreased by approximately 8.4%, 6.9%, 6.9% respectively. The RMSE of image registration RMSEreg decreased by 33.8%. In addition, the speed of the nursery tray image mosaic technology could meet the requirements for real-time implementation in super hybrid rice automated sowing machines.

show abstract

“…2) SURF descriptor The SURF descriptor is obtained by calculating the main direction and feature vector [21] . x y x y…”

Section: Image Mosaic Methodologies 221 the Phase Correlation Algormentioning

confidence: 99%

A fast nursery plug tray images sequence mosaic technique

Tan

et al. 2018

International Journal of Agricultural and Biological Engineering

View full text Add to dashboard Cite

show abstract

“…Therefore, the interest points, which includes their locations and scales, will be detected in approximate Gaussian scale space [25].…”

Section: Feature Matchingmentioning

confidence: 99%

“…For matching the features of those two images, we used the nearest neighbour method similar to [25]. In this way, image CSA uv has n 1 directed line segments and image CDE uv has n 2 directed line segments, and the nearest neighbour pair can be obtained by defining matrix K as below:…”

Section: Feature Matchingmentioning

confidence: 99%

Enhanced 3D Point Cloud from a Light Field Image

et al. 2020

View full text Add to dashboard Cite

The importance of three-dimensional (3D) point cloud technologies in the field of agriculture environmental research has increased in recent years. Obtaining dense and accurate 3D reconstructions of plants and urban areas provide useful information for remote sensing. In this paper, we propose a novel strategy for the enhancement of 3D point clouds from a single 4D light field (LF) image. Using a light field camera in this way creates an easy way for obtaining 3D point clouds from one snapshot and enabling diversity in monitoring and modelling applications for remote sensing. Considering an LF image and associated depth map as an input, we first apply histogram equalization and histogram stretching to enhance the separation between depth planes. We then apply multi-modal edge detection by using feature matching and fuzzy logic from the central sub-aperture LF image and the depth map. These two steps of depth map enhancement are significant parts of our novelty for this work. After combing the two previous steps and transforming the point–plane correspondence, we can obtain the 3D point cloud. We tested our method with synthetic and real world image databases. To verify the accuracy of our method, we compared our results with two different state-of-the-art algorithms. The results showed that our method can reliably mitigate noise and had the highest level of detail compared to other existing methods.

show abstract

“…Images with any size can be computed easily. In this research, SURF features are defined for the renal cells segmented from the images [11].…”

Section:  Surf Featuresmentioning

confidence: 99%

Detection of Renal carcinoma in Ultrasound Images using HOG and SURF features

Thamizhvani*,

Chandrasekaran,

Dhivya

et al. 2019

IJRTE

View full text Add to dashboard Cite

Cancer can be defined as the abnormal growth of cells in any region of the human. Cancer cells possess a special property called metastasis that involves the movement of cells from one location to another location. Renal cancer is becoming predominant and there are different types. One among them is Renal cell carcinoma mainly occurs in the renal tubules. In this study, ultrasound images are considered for the detection of renal cell carcinoma. The images undergo pre-processing to remove speckle noises. The region of interest is defined using region growing technique. Later Feature descriptors like histogram of oriented gradient features and speeded up robust features are extracted from the segmented region for the analysis of carcinoma. Texture features are also derived along with these descriptors. These features are classified using Adaptive Support Vector Machine for the diagnosis of the renal cell carcinoma from normal images. With the performance of the classifier, it is defined that feature descriptors illustrate the region of carcinoma more effectively

show abstract

Image mosaicking using SURF features of line segments

Cited by 20 publications

References 31 publications

A fast nursery plug tray images sequence mosaic technique

A fast nursery plug tray images sequence mosaic technique

Enhanced 3D Point Cloud from a Light Field Image

Detection of Renal carcinoma in Ultrasound Images using HOG and SURF features

Contact Info

Product

Resources

About