An FPGA-based accelerator for multiple real-time template matching

Albuquerque, Erika S.; Ferreira, Antonyus P. A.; Silva, Joao G. M.; Barbosa, João Paulo; Carlos, Renato L. M.; Albuquerque, Djeefther S.; Barros, Edna

doi:10.1109/sbcci.2016.7724071

Cited by 6 publications

(6 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to large amounts of MAC computations in (1), ZNCC formula should be simplified further considering stream processing and parallel computation on an FPGA. The simplified equation [25] is formulated and shown in Fig. 2.…”

Section: Zncc-based Template Matching On Fpgas a Zncc-based Template Matchingmentioning

confidence: 99%

“…It is obvious that template matching in a size of 32 × 32 runs once after 32 clock cycles. This kind of concurrent operation makes template matching run very fast, which is different from the memory-based approach [25]. But there is a shortcoming of low matching-based location precision at some position due to template size step in the column-scanning process along the row directions but high location precision due to one-pixel step along the column direction.…”

Section: B Image Caching On Fpgasmentioning

confidence: 99%

“…In order to evaluate our method with other competitive FPGA-based ZNCC method [25], we compare their hardware resource consumptions on a same Stratix IV FPGA (EP4SE530H35C2) used in [25]. The results are listed in Table 4.…”

Section: Table 2 Time Consumption Comparison (Unit:ms)mentioning

confidence: 99%

“…The results are listed in Table 4. Obviously our method consumes fewer hardware resources than the ZNCC circuit in [25] under non-DSP and DSP modes. The reasons are that the correlation window computations are implemented by MAC circuit in our designed circuit but feedback FIFOs in [25]'s circuit, and windowed image is projected by register-based pipelining technology in our method but by memory-based pipelining technology.…”

Section: Table 2 Time Consumption Comparison (Unit:ms)mentioning

confidence: 99%

“…Resource consumption comparisons between concurrent and our presented ZNCC circuits. Resource consumption comparisons between the ZNCC circuits by our method and[25]'s method.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Eye-to-Hand Robotic Visual Tracking Based on Template Matching on FPGAs

Chen

Zhang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

This paper presents a delta robotic visual-servoing tracking method using zero-mean normalized cross-correlation (ZNCC)-based grayscale template matching hardware core on field-programmable gate arrays (FPGAs). The concurrent FPGA-based ZNCC hardware core with cascading multiplicationaccumulate (MAC) circuits is designed, which can largely reduce FPGA hardware resource consumptions. A compact optical imaging system with a front 45 • -slant mirror and an optical filter film are proposed, which can efficiently filter out the background cluttered artifacts. The trajectory visual-servoing tracking and dynamic tracking experiments based on our built-up delta robotic visual tracking platform are implemented. The experimental results indicate that the presented FPGA-based embedded robotic visual tracking method can efficiently improve an object trajectory tracking performance.INDEX TERMS Robotic visual tracking, template matching, FPGA.

show abstract

Section: Zncc-based Template Matching On Fpgas a Zncc-based Template Matchingmentioning

confidence: 99%

Section: B Image Caching On Fpgasmentioning

confidence: 99%

Section: Table 2 Time Consumption Comparison (Unit:ms)mentioning

confidence: 99%

Section: Table 2 Time Consumption Comparison (Unit:ms)mentioning

confidence: 99%

“…Resource consumption comparisons between concurrent and our presented ZNCC circuits. Resource consumption comparisons between the ZNCC circuits by our method and[25]'s method.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Eye-to-Hand Robotic Visual Tracking Based on Template Matching on FPGAs

Chen

Zhang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

show abstract

FPGA-Based Vision Processing System for Automatic Online Player Tracking in Indoor Sports

Ibraheem

Irwansyah

Hagemeyer

et al. 2018

J Sign Process Syst

View full text Add to dashboard Cite

A Novel Parallel Architecture for Template Matching based on Zero-Mean Normalized Cross-Correlation

Wang

Han³

2019

IEEE Access

View full text Add to dashboard Cite

Template matching based on zero-mean normalized cross-correlation measure (ZNCC) has been widely used in a broad range of image processing applications. To meet the requirements for high processing speed, small size, and variable image size in automatic target recognition systems, a novel field-programmable gate array (FPGA)-based parallel architecture is presented in this paper for the ZNCC computation. The proposed architecture employs two groups of RAM blocks, one of which is used for the multiply-accumulate operations of the real and the reference images and the other for data rearrangement of the reference image, and their functions are switched through 2-input multiplexers when searching at the next row. Moreover, the sum of the pixels in the searching area of the real image is computed through serially accumulating the differences between the new column in the current searching area and the old column in the last searching area using one dual-port RAM. Simultaneously, the sum of the squares of the pixels is calculated in the same way. Using the Altera Stratix II FPGA chip (EP2S90F780I4) as the target device, the compilation results with Quartus II show that compared with the traditional architecture, the synthesis logic utilization decreases from 63% to 35% and the usage of DSP blocks decreases from 59% to 39%, while the memory bits only increase by 8% and the usage of other resources is nearly the same. The simulation and practical experimental results show that the proposed architecture can effectively improve the performance of the practical automatic target recognition system. INDEX TERMS FPGA, normalized cross-correlation measure, parallel architecture, template matching.

show abstract

An FPGA-based accelerator for multiple real-time template matching

Cited by 6 publications

References 13 publications

Eye-to-Hand Robotic Visual Tracking Based on Template Matching on FPGAs

Eye-to-Hand Robotic Visual Tracking Based on Template Matching on FPGAs

FPGA-Based Vision Processing System for Automatic Online Player Tracking in Indoor Sports

A Novel Parallel Architecture for Template Matching based on Zero-Mean Normalized Cross-Correlation

Contact Info

Product

Resources

About