Estimation of general 2D affine motion using Fourier descriptors

Oirrak, Ahmed El; Daoudi, Mohamed; Aboutajdine, Driss

doi:10.1016/s0031-3203(01)00019-x

Cited by 23 publications

(14 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, the scale range of 0.8 to 1.2 has been studied with a smaller step of 0.01. FDIM [24] however, produces better results but has shown a sharp decline of performance. These schemes produce fairly accurate matching statistics in case of no scaling i.e.…”

Section: Template Scale Analysismentioning

confidence: 97%

“…Robust image matching algorithm makes DT of both the main image and the template in order to reduce the computation in the main matching instructions loop [21], [22]. A similar attempt to find a better variant is to combine the DT with a normalized gradient consistency score [23], [24]. All these schemes search the required template features in the main image in a serial order which makes these schemes computationally expensive and sensitive to even slight scaling and rotation.…”

Section: Copyright C 2009 the Institute Of Electronics Information Amentioning

confidence: 99%

“…We have also compared the processing time of the proposed algorithm with the classical correlation scheme [10], NCC [13], successive similarity detection algorithm (SSDA) [15], Chamfer image matching algorithm [18], RIMA [21] and FDIM [24]. Results of this comparison are provided in Fig.…”

Section: Speed Analysismentioning

confidence: 99%

“…Thus, the adaptive prediction mechanism supervises the subset extraction process and provides a tight bound in the search area boundaries. The proposed algorithm is compared with six different algorithms, classical correlation scheme [10], NCC [13], SSDA [15], Chamfer image matching [18], Robust image matching algorithm (RIMA) [21] and Fourier Descriptors Image Matching (FDIM) [24]. The classical correlation scheme [10] accumulates pixel errors on the basis of their gray value difference.…”

Section: Adaptive Convergencementioning

confidence: 99%

“…The results were consolidated in Fig. 17 for the proposed algorithm as well as the other schemes [10], [13], [15], [18], [21], [24]. These results are provided in the percentage P of right prediction of the templates for each scale factor.…”

Section: Template Scale Analysismentioning

confidence: 99%

See 4 more Smart Citations

Visual Aerial Navigation through Adaptive Prediction and Hyper-Space Image Matching

Manzar

Cheema

Jalil

et al. 2009

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYImage matching is an important area of research in the field of artificial intelligence, machine vision and visual navigation. This paper presents a new image matching scheme suitable for visual navigation. In this scheme, gray scale images are sliced and quantized to form sub-band binary images. The information in the binary images is then signaturized to form a vector space and the signatures are sorted as per significance. These sorted signatures are then normalized to transform the represented image pictorial features in a rotation and scale invariant form. For the image matching these two vector spaces from both the images are compared in the transformed domain. This comparison yields efficient results directly in the image spatial domain avoiding the need of image inverse transformation. As compared to the conventional correlation, this comparison avoids the wide range of square error calculations all over the image. In fact, it directly guides the solution to converge towards the estimate given by the adaptive prediction for a high speed performance in an aerial video sequence. A four dimensional solution population scheme has also been presented with a matching confidence factor. This factor helps in terminating the iterations when the essential matching conditions have been achieved. The proposed scheme gives robust and fast results for normal, scaled and rotated templates. Speed comparison with older techniques shows the computational viability of this new technique and its much lesser dependence on image size. The method also shows noise immunity at 30 dB AWGN and impulsive noise.

show abstract

Section: Template Scale Analysismentioning

confidence: 97%