A novel rotation-invariant template matching based on HOG and AMDF for industrial laser cutting applications

Arslan, Osman; Demirci, B.; Altun, Halis; Tunaboylu, N.S.

doi:10.1109/isma.2013.6547367

Cited by 3 publications

(2 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Traditional template matching generates the similarity map based on pixel intensity values, and is not robust to hard matching scenarios such as significant non-rigid deformations of the object, changes in the illumination and size of the target, and occlusion [7]. To address this issue, more distinctive hand-crafted features such as scale-invariant feature transform (SIFT) [8] and histogram of oriented gradients (HOG) [9] can be used instead of the intensity values for robust template matching [10][11][12][13]. However, these features must be extracted by certain manually predefined algorithms based on expert knowledge, and therefore have limited description capabilities [14].…”

Section: Introductionmentioning

confidence: 99%

Shape–Texture Debiased Training for Robust Template Matching

Gao

Spratling

2022

Sensors

View full text Add to dashboard Cite

Finding a template in a search image is an important task underlying many computer vision applications. This is typically solved by calculating a similarity map using features extracted from the separate images. Recent approaches perform template matching in a deep feature space, produced by a convolutional neural network (CNN), which is found to provide more tolerance to changes in appearance. Inspired by these findings, in this article we investigate whether enhancing the CNN’s encoding of shape information can produce more distinguishable features that improve the performance of template matching. By comparing features from the same CNN trained using different shape–texture training methods, we determined a feature space which improves the performance of most template matching algorithms. When combining the proposed method with the Divisive Input Modulation (DIM) template matching algorithm, its performance is greatly improved, and the resulting method produces state-of-the-art results on a standard benchmark. To confirm these results, we create a new benchmark and show that the proposed method outperforms existing techniques on this new dataset.

show abstract

Section: Introductionmentioning

confidence: 99%

Shape–Texture Debiased Training for Robust Template Matching

Gao

Spratling

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Although a similar line of studies is reported in the literature [5,9,13], which * Correspondence: mpeker@nigde.edu.tr employ HOG for shape detection, we are proposing a new hardware-friendly algorithm, which employs AMDF as an effective and easy-to-implement decision module on FPGA. The method uses a set of features extracted from the HOG module, which is later classified by the AMDF module as explained in [14][15][16]. In this paper, we will demonstrate the effective implementation of the proposed method on FPGA for real-time applications.…”

Section: Introductionmentioning

confidence: 99%

Hardware implementation of a scale and rotation invariant object detection algorithm on FPGA for real-time applications

Peker¹,

Altun²,

Karakaya³

2016

Turk J Elec Eng & Comp Sci

Self Cite

View full text Add to dashboard Cite

Abstract:A hardware implementation of a computationally light, scale, and rotation invariant method for shape detection on FPGA is devised. The method is based on histogram of oriented gradients (HOG) and average magnitude difference function (AMDF). AMDF is used as a decision module that measures the similarity/dissimilarity between HOG vectors of an image in order to classify the object. In addition, a simulation environment implemented on MATLAB is developed in order to overcome the time-consuming and tedious process of hardware verification on the FPGA platform.The simulation environment provides specific tools to quickly implement the proposed methods. It is shown that the simulator is able to produce exactly the same results as those obtained from FPGA implementation. The results indicate that the proposed approach leads to a shape detection method that is computationally light, scale, and rotation invariant, and, therefore, suitable for real-time industrial and robotic vision applications.

show abstract