Enhanced OCR Based on Deep Learning Technology
The technologies for optical character recognition (OCR) have been further improved. MERLIC now comes with an OCR classifier that is based on deep learning and can be applied to a wide range of fonts. This new function results in higher-than-ever detection rates for number and character combinations, e.g., on workpieces, for reliable identification and further processing. Additionally, dot-print fonts are now read more robustly.
Improved Bar and Data Code Reading
The reading of bar and data codes has been improved: Blurry, over-exposed, distorted, or low-contrast QR codes are read more robustly, and QR codes with uneven column widths can be read without problems. Furthermore, MERLIC can also read partially occluded or partially defective bar codes. The recognition of ECC 200 codes with missing finder patterns as well as the robustness against false positives were also optimized.
An improvement in performance for the alignment and reading of texts has been achieved: The alignment of an image will now execute up to 124 percent faster. Also, the speed for reading texts and numbers for non-dot-print fonts has been improved by about 224 percent.
Several further tool improvements could be achieved with this release: For example, all matching tools in MERLIC now support correlation-based matching. Additionally, several filtering tools as well as the usability of, e.g., 'Fit Circle', have been improved.
New ADLINK I/O Interfaces
MERLIC supports the digital I/Os of the ADLINK NEON 1020/1040 smart cameras and the EOS 1300 vision system.
Since the last product version we have been working hard to improve MERLIC in every way. This has, however, resulted in the following incompatibility. We apologize for the inconvenience.
- The MERLIC tool procedure 'Me_get_example_dir' now returns the file path in UTF-8 instead of the native encoding. Therefore, existing MERLIC Vision Apps which use images from the example directory need to be adjusted to use the updated path. More information.
- The following tools are no longer compatible since they have been extended by the result 'Processed Region':
- 'Adapt Brightness'
- 'Apply Gray Morphology'
- 'Correct Shading'
- 'Emphasize Edges'
- 'Merge Images'
- 'Reduce Image Noise'
- The following tools are no longer compatible since the parameters 'Processing Region' and 'ROI' have been removed:
- 'Align Image'
- 'Convert Pixel Type'
- 'Transform Color Space'
- 'Transform to Grayscale Image'