Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Image histogram equalization by the image input connector #778

Closed
beniz opened this issue Sep 1, 2020 · 0 comments · Fixed by #782 or #805
Closed

Image histogram equalization by the image input connector #778

beniz opened this issue Sep 1, 2020 · 0 comments · Fixed by #782 or #805
Assignees
@sileht @beniz
Labels
datatype:images kind:API kind:connectors

Comments

@beniz
Copy link
Collaborator

beniz commented Sep 1, 2020 •
edited
Loading

Having OpenCV's histogram equalization directly executed by the input connector is useful in some cases for both training and inference.

We suggest a eqhist boolean parameter to be added to the input JSON object on both training and predict calls.

I guess this could go after the rgb check here: https://github.com/jolibrain/deepdetect/blob/master/src/imginputfileconn.h#L104

The function to apply from OpenCV should be:

/// Apply Histogram Equalization
cv::equalizeHist( src, dst );

Beware that equalization is per channel, so the call above may only work when the image is single channel B&W, which is our motivating use case actually. For RGB images, the equalization may need to be applied per channel.
@beniz beniz assigned beniz and sileht Sep 1, 2020
sileht added a commit to sileht/deepdetect that referenced this issue Sep 7, 2020
@sileht
feat(imginputfile): histogram equalization of input image
790e78b 
This does a histogram equalization of colored image.

Closes jolibrain#778
@sileht sileht mentioned this issue Sep 7, 2020
Merged
sileht added a commit to sileht/deepdetect that referenced this issue Sep 7, 2020
@sileht
feat(imginputfile): histogram equalization of input image
027dc84 
This does a histogram equalization of bw & colored image.

Closes jolibrain#778
sileht added a commit to sileht/deepdetect that referenced this issue Sep 8, 2020
@sileht
feat(imginputfile): histogram equalization of input image
4a3ef29 
This does a histogram equalization of bw & colored image.

Closes jolibrain#778
sileht added a commit to sileht/deepdetect that referenced this issue Sep 8, 2020
@sileht
feat(imginputfile): histogram equalization of input image
b8928cb 
This does a histogram equalization of bw & colored image.

Closes jolibrain#778
sileht added a commit to sileht/deepdetect that referenced this issue Sep 8, 2020
@sileht
feat(imginputfile): histogram equalization of input image
99fd282 
This does a histogram equalization of bw & colored image.

Closes jolibrain#778
sileht added a commit that referenced this issue Sep 10, 2020
@sileht
feat(imginputfile): histogram equalization of input image
576f2d8 
This does a histogram equalization of bw & colored image.

Closes #778
sileht added a commit to sileht/deepdetect that referenced this issue Sep 10, 2020
@sileht
feat(imginputfile): histogram equalization of input image
fd93b68 
This does a histogram equalization of bw & colored image.

Closes jolibrain#778
@sileht sileht mentioned this issue Sep 10, 2020
Closed
sileht added a commit to sileht/deepdetect that referenced this issue Sep 14, 2020
@sileht
feat(imginputfile): histogram equalization of input image
9fa9dbc 
This does a histogram equalization of bw & colored image.

Closes jolibrain#778
sileht added a commit to sileht/deepdetect that referenced this issue Sep 14, 2020
@sileht
feat(imginputfile): histogram equalization of input image
cda1e65 
This does a histogram equalization of bw & colored image.

Closes jolibrain#778
@sileht sileht mentioned this issue Sep 14, 2020
Merged
sileht added a commit to sileht/deepdetect that referenced this issue Sep 17, 2020
@sileht
feat(imginputfile): histogram equalization of input image
f0b082a 
This does a histogram equalization of bw & colored image.

Closes jolibrain#778
sileht added a commit to sileht/deepdetect that referenced this issue Sep 17, 2020
@sileht
feat(imginputfile): histogram equalization of input image
9985327 
This does a histogram equalization of bw & colored image.

Closes jolibrain#778
sileht added a commit to sileht/deepdetect that referenced this issue Sep 17, 2020
@sileht
feat(imginputfile): histogram equalization of input image
f5ad8de 
This does a histogram equalization of bw & colored image.

Closes jolibrain#778
sileht added a commit to sileht/deepdetect that referenced this issue Sep 18, 2020
@sileht
feat(imginputfile): histogram equalization of input image
eab8690 
This does a histogram equalization of bw & colored image.

Closes jolibrain#778
sileht added a commit that referenced this issue Sep 18, 2020
@sileht
feat(imginputfile): histogram equalization of input image
2f0061c 
This does a histogram equalization of bw & colored image.

Closes #778
sileht pushed a commit that referenced this issue Oct 5, 2020
@semantic-release-bot
chore(release): 0.10.0 [skip ci]
b1346d8 
# [0.10.0](v0.9.7...v0.10.0) (2020-10-05)

### Bug Fixes

* missing variant package in docker files ([dcf738b](dcf738b))
* **build:** CUDA_ARCH not escaped correctly ([696087f](696087f))
* in tensorrt builds, remove forced cuda version and unused lib output + force-select tensorrt when tensorrt_oss is selected ([9430fb4](9430fb4))
* **clang-format:** signed/unsigned comparaison ([af8e144](af8e144))
* **clang-format:** signed/unsigned comparaison ([0ccabb6](0ccabb6))
* **dede:** Remove unnecessary caffe include that prevent build with torch only ([a471b82](a471b82))
* **docker:** install rapidjson-dev package ([30fb2ca](30fb2ca))
* **native:** do not raise exception if no template_param is given ([d0705ab](d0705ab))
* **nbeats:** correctly setup trend and seasonality models (implement paper version and not code version) ([75accc6](75accc6))
* **nbeats:** much lower memory use in case of large dim signals ([639e222](639e222))
* **tests:** inc iteration of torchapi.service_train_image test ([4c93ace](4c93ace))
* /api/ alias when deployed on deepdetect.com ([4736893](4736893))
* add support and automated processing of categorical variables in timeseries data ([1a9af3e](1a9af3e))
* allow serialization/deserializationt of Inf/-Inf/NaN ([976c892](976c892))
* allows to specify size and color/bw with segmentation models ([58ecb4a](58ecb4a))
* build with -DUSE_TENSORRT_OSS=ON ([39bd675](39bd675))
* convolution layer initialization of SE-ResNeXt network templates ([69ff0fb](69ff0fb))
* input image transforms in API doc ([f513f17](f513f17))
* install cmake version 3.10 ([10666b8](10666b8))
* race condition in xgboost|dede build ([fd32eae](fd32eae))
* replace db":true by db":false in json files when copying models ([06ac6df](06ac6df))
* set caffe smooth l1 loss threshold to 1 ([0e329f0](0e329f0))
* ssd_300_res_128 deploy file is missing a quote ([4e52a0e](4e52a0e))
* svm prediction with alll db combinations ([d8e2961](d8e2961))
* svm with db training ([6e925f2](6e925f2))
* tensorrt does not support blank_label ([7916500](7916500))
* update caffe cudnn engine without template ([ca58c51](ca58c51))
* **build:** ensure all xgboost submodules are checkouted ([12aaa1a](12aaa1a))
* **ci:** add CUDA_ARCH ([5b9eb15](5b9eb15))
* **ci:** add Jenkingfile symlink for old PR ([61e1176](61e1176))
* **ci:** add missing LD_LIBRARY_PATH for protoc ([cae5b1b](cae5b1b))
* **ci:** add missing steps block ([549bc59](549bc59))
* **ci:** cleanup workspace after artefact upload ([c9321cd](c9321cd))
* **ci:** fix dataset path location ([d071ff3](d071ff3))
* **ci:** fix deepdetect-pytorch project name ([0857b48](0857b48))
* **ci:** remove empty stage ([a7f56e4](a7f56e4))
* **ci:** run clang-format from build directory ([c51a022](c51a022))
* **ci:** typo in Jenkinfile ([a512f5b](a512f5b))
* **ci:** use correct cmake TORCH option ([1553327](1553327))
* **ci:** use unsuccessful instead of failure ([5dc1571](5dc1571))
* **clang-format:** typo in dataset tarball command ([04ddad7](04ddad7))
* **dede:** support all version of spdlog while building with syslog ([81f47c9](81f47c9))
* **torch:** handle case where sequence data is < wanted timestep ([b6d394a](b6d394a))
* **TRT:** refinedet ([b6152b6](b6152b6))
* unusual builds (ie w/o torch or with tsne only lead to build errors ([241bf6b](241bf6b))

### Features

* **build:** add script to create cppnet-lib debian package ([28247b4](28247b4))
* **build:** allow to change CUDA_ARCH ([67ad43e](67ad43e))
* **ci:** add cache for examples/ dataset ([4c19d78](4c19d78))
* **ci:** add new jenkins job to cache pytorch build ([43c9c06](43c9c06))
* **ci:** enable ccache ([c72327d](c72327d))
* **ci:** lock the GPU for running tests ([dd28248](dd28248))
* **ci:** only build last version of a branch ([53edf52](53edf52))
* **ci:** use prebuilt dataset for unittests ([28ab168](28ab168))
* **dede:** Training for image classification with torch ([6e81915](6e81915))
* **graph:** lstm autoencoder ([038a74c](038a74c))
* **imginputfile:** histogram equalization of input image ([2f0061c](2f0061c)), closes [#778](#778)
* **stats:** added service statistics mechanism ([1839e4a](1839e4a))
* **torch:** in case of timeseries, warn if file do not contain enough timesteps ([1a5f905](1a5f905))
* **torch:** nbeats ([f288665](f288665))
* **torch:** upgrade to torch 1.6 ([f8f7dbb](f8f7dbb))
* **torch,native:** extract_layer ([d37e182](d37e182))
* add json output to dd_bench.py ([874fc01](874fc01))
* added bw image input support to dd_bench ([6e558d6](6e558d6))
* **trains-status:** add tflops to body.measures ([af31c8b](af31c8b)), closes [#785](#785)
* Docker images optimization ([fba637a](fba637a))
* format the code with clang-format ([07d6bdc](07d6bdc))
* LSTM over torch , preliminary internal graph representation ([25faa8b](25faa8b))
* update all docker images to ubuntu 18.04 ([eaf0421](eaf0421))
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@sileht sileht

@beniz beniz

Labels
datatype:images kind:API kind:connectors
Projects
None yet
Milestone
No milestone
2 participants
@sileht @beniz