Detecting objects using openCV and sift
from a_cv_sift_detection import SiftMatchingOnScreen
needle_images = [
r"C:\detectiontest\media_manager_icon--744x194--1250x738.png",
r"C:\detectiontest\chrome_icon--643x199--1140x734.png",
r"C:\detectiontest\einstellungen_icon--537x200--1038x735.png",
r"C:\detectiontest\kamera_icon--426x203--931x738.png",
r"C:\detectiontest\spiele_und_gewinne_icon--1347x0--1920x452.png",
r"C:\detectiontest\bluestacks_x_icon--1101x0--1643x449.png",
r"C:\detectiontest\roblox_icon--833x0--1342x448.png",
r"C:\detectiontest\systemapps_icon--528x0--1067x448.png",
r"C:\detectiontest\gamecenter_icon--244x0--781x451.png",
r"C:\detectiontest\playstore_icon--0x0--478x451.png",
]
siftdetect = SiftMatchingOnScreen()
siftdetect.configure_monitor(monitor=1)
siftdetect.get_needle_images(needle_images, scale_percent=100)
while True:
siftdetect.get_screenshot_and_start_detection(
checks=50,
trees=5,
debug=False,
max_distance=100,
minimum_matches_per_group=5,
show_results=True,
sleep_time_for_results=0.1,
quit_key_for_results="q",
scale_percent=100,
)
sleep(5)
while True:
siftdetect.get_screenshot_and_start_detection(
checks=50,
trees=5,
debug=False,
max_distance=100,
minimum_matches_per_group=5,
show_results=False,
sleep_time_for_results=0.1,
quit_key_for_results="q",
scale_percent=100,
)
print(siftdetect.df)
0 956.575073 ... C:\detectiontest\media_manager_icon--744x194--...
1 960.470520 ... C:\detectiontest\media_manager_icon--744x194--...
2 961.128601 ... C:\detectiontest\media_manager_icon--744x194--...
3 964.380310 ... C:\detectiontest\media_manager_icon--744x194--...
4 965.097656 ... C:\detectiontest\media_manager_icon--744x194--...
.. ... ... ...
420 257.061249 ... C:\detectiontest\playstore_icon--0x0--478x451.png
421 263.247589 ... C:\detectiontest\playstore_icon--0x0--478x451.png
422 263.265106 ... C:\detectiontest\playstore_icon--0x0--478x451.png
423 263.682190 ... C:\detectiontest\playstore_icon--0x0--478x451.png
424 267.074219 ... C:\detectiontest\playstore_icon--0x0--478x451.png
[425 rows x 7 columns]
x ... needle
0 762.877319 ... C:\detectiontest\media_manager_icon--744x194--...
1 781.889160 ... C:\detectiontest\media_manager_icon--744x194--...
2 816.500549 ... C:\detectiontest\media_manager_icon--744x194--...
3 816.339050 ... C:\detectiontest\media_manager_icon--744x194--...
4 816.586853 ... C:\detectiontest\media_manager_icon--744x194--...
.. ... ... ...
344 257.061249 ... C:\detectiontest\playstore_icon--0x0--478x451.png
345 263.247589 ... C:\detectiontest\playstore_icon--0x0--478x451.png
346 263.265106 ... C:\detectiontest\playstore_icon--0x0--478x451.png
347 263.682190 ... C:\detectiontest\playstore_icon--0x0--478x451.png
348 267.074219 ... C:\detectiontest\playstore_icon--0x0--478x451.png
[349 rows x 7 columns]
No objects to concatenate
x ... needle
0 881.638245 ... C:\detectiontest\media_manager_icon--744x194--...
1 935.431885 ... C:\detectiontest\media_manager_icon--744x194--...
2 955.275879 ... C:\detectiontest\media_manager_icon--744x194--...
3 941.385254 ... C:\detectiontest\media_manager_icon--744x194--...
4 817.367920 ... C:\detectiontest\media_manager_icon--744x194--...
.. ... ... ...
302 392.155945 ... C:\detectiontest\playstore_icon--0x0--478x451.png
303 428.587921 ... C:\detectiontest\playstore_icon--0x0--478x451.png
304 431.010773 ... C:\detectiontest\playstore_icon--0x0--478x451.png
305 476.075165 ... C:\detectiontest\playstore_icon--0x0--478x451.png
306 323.301575 ... C:\detectiontest\playstore_icon--0x0--478x451.png
[307 rows x 7 columns]
x ... needle
0 881.638245 ... C:\detectiontest\media_manager_icon--744x194--...
1 937.933167 ... C:\detectiontest\media_manager_icon--744x194--...
2 935.431885 ... C:\detectiontest\media_manager_icon--744x194--...
3 955.450562 ... C:\detectiontest\media_manager_icon--744x194--...
4 941.385254 ... C:\detectiontest\media_manager_icon--744x194--...
.. ... ... ...
303 417.567749 ... C:\detectiontest\playstore_icon--0x0--478x451.png
304 429.982422 ... C:\detectiontest\playstore_icon--0x0--478x451.png
305 438.710541 ... C:\detectiontest\playstore_icon--0x0--478x451.png
306 445.110016 ... C:\detectiontest\playstore_icon--0x0--478x451.png
307 458.563171 ... C:\detectiontest\playstore_icon--0x0--478x451.png
[308 rows x 7 columns]
x ... needle
0 996.570251 ... C:\detectiontest\media_manager_icon--744x194--...
1 1011.908936 ... C:\detectiontest\media_manager_icon--744x194--...
2 1015.692566 ... C:\detectiontest\media_manager_icon--744x194--...
3 1056.677734 ... C:\detectiontest\media_manager_icon--744x194--...
4 1071.151245 ... C:\detectiontest\media_manager_icon--744x194--...
.. ... ... ...
345 257.061249 ... C:\detectiontest\playstore_icon--0x0--478x451.png
346 263.247589 ... C:\detectiontest\playstore_icon--0x0--478x451.png
347 263.265106 ... C:\detectiontest\playstore_icon--0x0--478x451.png
348 263.682190 ... C:\detectiontest\playstore_icon--0x0--478x451.png
349 267.074219 ... C:\detectiontest\playstore_icon--0x0--478x451.png
[350 rows x 7 columns]
x ... needle
0 996.570251 ... C:\detectiontest\media_manager_icon--744x194--...
1 1011.908936 ... C:\detectiontest\media_manager_icon--744x194--...
2 1015.692566 ... C:\detectiontest\media_manager_icon--744x194--...
3 1056.677734 ... C:\detectiontest\media_manager_icon--744x194--...
4 1071.151245 ... C:\detectiontest\media_manager_icon--744x194--...
.. ... ... ...
352 257.061249 ... C:\detectiontest\playstore_icon--0x0--478x451.png
353 263.247589 ... C:\detectiontest\playstore_icon--0x0--478x451.png
354 263.265106 ... C:\detectiontest\playstore_icon--0x0--478x451.png
355 263.682190 ... C:\detectiontest\playstore_icon--0x0--478x451.png
356 267.074219 ... C:\detectiontest\playstore_icon--0x0--478x451.png
[357 rows x 7 columns]
x ... needle
0 762.877319 ... C:\detectiontest\media_manager_icon--744x194--...
1 781.889160 ... C:\detectiontest\media_manager_icon--744x194--...
2 816.500549 ... C:\detectiontest\media_manager_icon--744x194--...
3 816.339050 ... C:\detectiontest\media_manager_icon--744x194--...
4 816.586853 ... C:\detectiontest\media_manager_icon--744x194--...
.. ... ... ...
343 257.061249 ... C:\detectiontest\playstore_icon--0x0--478x451.png
344 263.247589 ... C:\detectiontest\playstore_icon--0x0--478x451.png
345 263.265106 ... C:\detectiontest\playstore_icon--0x0--478x451.png
346 263.682190 ... C:\detectiontest\playstore_icon--0x0--478x451.png
347 267.074219 ... C:\detectiontest\playstore_icon--0x0--478x451.png
[348 rows x 7 columns]
Metadata
Release files for a-cv-sift-detection 0.10
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| a_cv_sift_detection-0.10.tar.gz | 9.8 kB | Without provenance | Details for a_cv_sift_detection-0.10.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| a_cv_sift_detection-0.10-py3-none-any.whl | Python 3 | none | any | Details for a_cv_sift_detection-0.10-py3-none-any.whl |
Total release size: 20.4 kB
Release files / a_cv_sift_detection-0.10.tar.gz
| Download URL | a_cv_sift_detection-0.10.tar.gz |
|---|---|
| Size | 9.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
7e6fb1088a7336492a8e3e8675167a37d0294784c26d605f98d330789ff2bcb8
|
|
BLAKE2b-256 checksum How to use checksums |
3b6e669bdd5ce1fb00f7804af0007cc3f5c742d9065f6aee80820c07e4de559a
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.1 CPython/3.9.13
|
Release files / a_cv_sift_detection-0.10-py3-none-any.whl
| Download URL | a_cv_sift_detection-0.10-py3-none-any.whl |
|---|---|
| Size | 10.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
0d3cbd70cf36768b384d852b27a199e774f27091305c9c119cdac5768c89841e
|
|
BLAKE2b-256 checksum How to use checksums |
d9f57a8114987b2db40eba83f19efda688e7f454d1c7a9e5ebe92490e06c76a4
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.1 CPython/3.9.13
|
Security for a-cv-sift-detection 0.10
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.