v1.4.0

v1.4.0 is a foundational imlib reorganization. find_blobs() / find_markers() were rewritten for direct color-threshold tracking, many standalone image methods were folded into blob tuples, and several image methods were renamed (subimg → copy, blit → replace, find_eyes → find_eye, compress → compressed). It also adds sensor.get_fb() / AGC / AEC toggles and ~60 example scripts. This is a heavily breaking release — read the breaking changes below.

Highlights

  • Color-based blob tracking — find_blobs() takes color thresholds directly and returns a flat list of blob tuples.

  • Imlib cleanup — subimg() → copy(), blit() → replace(), find_eyes() → find_eye(), compress() → compressed().

  • Sensor — added sensor.get_fb() and AGC/AEC enable toggles.

  • ~60 example scripts added/expanded.

  • Breaking: the blob API, removed image methods, and several renames — see the breaking changes.

New features

  • Sensor — added sensor.get_fb() (retrieve the last snapshot framebuffer), sensor.set_gain_ctrl(enable) (toggle AGC), and sensor.set_exposure_ctrl(enable) (toggle AEC).

  • Image — image subscript (img[i]) now supports both reading and writing pixels (grayscale / RGB565 / JPEG); added an in-place image.compress() and a quality= keyword on the JPEG functions; find_blobs() / find_markers() / find_features() now return an empty (iterable) list instead of a non-list when nothing is found.

  • Added ~60 example scripts (board control, image filters, color tracking, face / eye / feature detection, WiFi shield) and a second iris-detection example.

Other changes and improvements

  • image.histeq() now works on RGB565 images, not just grayscale.

Bug fixes

  • Initialized pyb.DAC on boot so the DAC works (#121), fixed a broken image.dilate(), the WINC client.send() zero-byte corruption, blob merge re-iteration, a file-buffer reinitialization bug that broke the filesystem after first use, the thermopile To_min / To_max computation (#113), and applied an upstream MicroPython HAL I2C/DMA patch.

Hardware and board support

  • WiFi shield — WINC1500 updated to Rev-B board files.

  • pyb.DAC — DAC peripheral enabled.

Breaking API changes

User-visible API breaks between v1.3.0 and v1.4.0. Scope: Python C-modules in modules/ and Python libraries in scripts/libraries/.

Each change is tagged with its impact:

  • major — affects most scripts that used the feature; you will need to port code.

  • minor — narrow API; only affects scripts that used it.

Changes are grouped by impact in that order. If you just want to port your code, jump to the migration checklist at the end. Each commit hash links to its diff on GitHub.

find_blobs() / find_markers() rewritten (major)

image.find_blobs() / find_markers() were rewritten for color-based tracking: you now pass color thresholds (and an optional filter function) directly instead of operating on a pre-segmented image, and the result is a single flat list of 10-element blob tuples (with a color-code field) instead of a nested per-color tree. Rework color-tracking scripts to pass color thresholds and iterate the flat list.

Commits: d1ff36602, 16daeb83a

Standalone image methods removed (major)

image.threshold, image.centroid(), image.orientation_radians(), image.orientation_degrees(), and image.pixels() were removed. The centroid / orientation / pixel-count data is now available as fields of the blob tuples returned by find_blobs().

Commits: 9e5d379c1

Imlib method renames (major)

Several image methods were renamed/reworked: image.subimg() → image.copy(), image.blit() → image.replace(), and image.blend() was changed to match the other two-image ops. The image.scale() / scaled() / rainbow() methods were removed (rainbow moved to the FIR module). Update calls to the new names.

Commits: b534dc6ff

find_eyes() renamed to find_eye() (minor)

image.find_eyes() was renamed to image.find_eye() and now returns a 2-tuple iris (x, y) instead of the previous eyes tuple. Rename the call and update the return handling.

Commits: 7a01a9bed

image.compress() renamed to compressed() (minor)

image.compress() (which returned a new JPEG image) was renamed to image.compressed(); the new image.compress() instead compresses the image in place and returns None. image.save() also gained a quality= keyword and now returns None instead of True. Migrate img.compress(q) (expecting a returned image) to img.compressed(quality=q).

Commits: 367e5fdd3

Migration checklist

For a clean port to v1.4.0 the typical work is:

  1. Rework color-tracking code to pass color thresholds to find_blobs() / find_markers() and iterate the flat blob-tuple list (the find_blobs rewrite).

  2. Read centroid / orientation / pixel data from the blob tuples instead of the removed standalone methods (the removed image methods).

  3. Rename subimg → copy, blit → replace, fix blend() calls, and replace scale / scaled / rainbow (the imlib renames).

  4. Rename find_eyes() → find_eye() (the find_eye rename) and compress() → compressed() (the compress rename).

All other scripts run unchanged.