Figure 1.
A Valid Indian Bank Cheque Image
Figure 2.
Automated Detection of an invalid Indian Bank Cheque Image
Table 1.
Challenges of extracting and recognizing data fields from bank cheque images
| Challenges | Difficulties |
|---|---|
| Data deterioration |
|
| The problem of Skewness |
|
| Distinct handwriting |
|
| Data superposition |
|
| Perplexity |
|
| Document torn and folded |
|
| Variation in image contrast |
|
| Cheque Streaks |
|
| Image compression |
|
| Integrated algorithm |
|
Figure 3.
Overview of the process flow of the proposed framework
Figure 4.
Overview of the Mask RCNN architecture
Figure 5.
Overview of the Resnet 101 architecture of the Mask RCNN Model
Figure 6.
Flow-diagram of First Stage validation error reporting module
Figure 7.
Annotated Bank cheque image for stage-1 Mask RCNN model
Figure 8.
Flow-diagram of Second Stage validation error reporting module
Figure 9.
Annotated Bank cheque image for stage-2 Mask RCNN model
Figure 10.
Different types of unacceptable overwritten/strike-through handwritten characters
Figure 11.
A few Bank Cheque sample images used for developing Stage-1 Mask RCNN Model
Figure 12.
A few Masked Bank Cheque sample images used for developing Stage-2 Mask RCNN Model
Table 2.
Summary of the Dataset
| Cheques with Missing Handwritten Field | Cheques with No Missing Handwritten Field | Cheques with Overwritten/Strike-through Characters | Cheques with no Overwritten/Strike-through Characters |
|---|---|---|---|
| 90 | 30 | 78 | 12 |
Figure 13.
Loss Graph During Training Phase of Mask RCNN Models
Table 3.
Qualitative Observations of Stage-1 Validation Module’s Performances
| U-Net | YOLO v8 | Proposed Mask RCNN based model |
|---|---|---|
 |  |  |
Table 5.
Qualitative Observations of Stage-2 Validation Module’s Performances
| U-Net | YOLO v8 | Proposed Mask RCNN based model |
|---|---|---|
 |  |  |
Table 6.
Quantitative Results: Detection Accuracy of Stage-2 Validation Module
| Overwritten/Strikethrough class types | U-Net based model (%) | YOLO v8 based model (%) | Proposed Mask R-CNN based model (%) |
|---|---|---|---|
| Type - A | 94.30 | 95.40 | 98.20 |
| Type – B | 93.90 | 94.56 | 98.03 |
| Type – C | 94.75 | 95.2 | 97.57 |
| Type – D | 95.39 | 95.42 | 98.50 |
| Type – E | 92.67 | 93.71 | 98.90 |
| Type – F | 93.25 | 95.83 | 97.82 |
| Type – G | 95.28 | 95.07 | 98.23 |
| Type – H | 92.33 | 94.89 | 97.90 |
| Type – I | 93.80 | 94.77 | 97.88 |