simon987/sist2
1
0
Fork 0
mirror of https://github.com/simon987/sist2.git synced 2025-04-19 18:26:43 +00:00
Code Issues Projects Releases Wiki Activity

OCR fixes

Browse Source
This commit is contained in:
simon987 2023-01-13 20:13:20 -05:00
parent a074d8cf10
commit b9f008603a
5 changed files with 16 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
.dockerignore
View File

@ -8,6 +8,7 @@ Testing/
**/cmake_install.cmake **/cmake_install.cmake
**/CMakeCache.txt **/CMakeCache.txt
**/CMakeFiles/ **/CMakeFiles/
.cmake
LICENSE LICENSE
Makefile Makefile
**/*.md **/*.md

6
Dockerfile
View File

@ -14,7 +14,7 @@ RUN cmake -DSIST_PLATFORM=x64_linux -DSIST_DEBUG=off -DBUILD_TESTS=off -DCMAKE_T
RUN make -j$(nproc) RUN make -j$(nproc)
RUN strip sist2 || mv sist2_debug sist2 RUN strip sist2 || mv sist2_debug sist2
FROM --platform="linux/amd64" ubuntu@sha256:b25ef49a40b7797937d0d23eca3b0a41701af6757afca23d504d50826f0b37ce FROM --platform="linux/amd64" ubuntu@sha256:965fbcae990b0467ed5657caceaec165018ef44a4d2d46c7cdea80a9dff0d1ea
WORKDIR /root WORKDIR /root
@ -33,7 +33,9 @@ RUN mkdir -p /usr/share/tessdata && \
curl -o /usr/share/tessdata/eng.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/eng.traineddata &&\ curl -o /usr/share/tessdata/eng.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/eng.traineddata &&\
curl -o /usr/share/tessdata/fra.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/fra.traineddata &&\ curl -o /usr/share/tessdata/fra.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/fra.traineddata &&\
curl -o /usr/share/tessdata/rus.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/rus.traineddata &&\ curl -o /usr/share/tessdata/rus.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/rus.traineddata &&\
curl -o /usr/share/tessdata/spa.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/spa.traineddata curl -o /usr/share/tessdata/osd.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/osd.traineddata &&\
curl -o /usr/share/tessdata/spa.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/spa.traineddata &&\
curl -o /usr/share/tessdata/chi_sim.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/chi_sim.traineddata
# sist2 # sist2
COPY --from=build /build/sist2 /root/sist2 COPY --from=build /build/sist2 /root/sist2

6
src/main.c
View File

@ -209,8 +209,10 @@ void initialize_scan_context(scan_args_t *args) {
ScanCtx.ebook_ctx.content_size = args->content_size; ScanCtx.ebook_ctx.content_size = args->content_size;
ScanCtx.ebook_ctx.enable_tn = args->tn_count > 0; ScanCtx.ebook_ctx.enable_tn = args->tn_count > 0;
ScanCtx.ebook_ctx.tn_size = args->tn_size; ScanCtx.ebook_ctx.tn_size = args->tn_size;
ScanCtx.ebook_ctx.tesseract_lang = args->tesseract_lang; if (args->ocr_ebooks) {
ScanCtx.ebook_ctx.tesseract_path = args->tesseract_path; ScanCtx.ebook_ctx.tesseract_lang = args->tesseract_lang;
ScanCtx.ebook_ctx.tesseract_path = args->tesseract_path;
}
ScanCtx.ebook_ctx.log = _log; ScanCtx.ebook_ctx.log = _log;
ScanCtx.ebook_ctx.logf = _logf; ScanCtx.ebook_ctx.logf = _logf;
ScanCtx.ebook_ctx.store = _store; ScanCtx.ebook_ctx.store = _store;

5
src/web/static_generated.c vendored
View File

File diff suppressed because one or more lines are too long

2
third-party/libscan/libscan/media/media.c vendored
View File

@ -446,7 +446,7 @@ int decode_frame_and_save_thumbnail(scan_media_ctx_t *ctx, AVFormatContext *pFor
return SAVE_THUMBNAIL_FAILED; return SAVE_THUMBNAIL_FAILED;
} }
if (ctx->tesseract_lang != NULL && IS_VIDEO(pFormatCtx) && thumbnail_index == 0) { if (ctx->tesseract_lang != NULL && thumbnail_index == 0) {
ocr_image(ctx, doc, decoder, frame_and_packet->frame); ocr_image(ctx, doc, decoder, frame_and_packet->frame);
} }