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

ocr-engin-test-style-fixes #55

Merged
merged 1 commit into from
Jun 15, 2024
Merged

ocr-engin-test-style-fixes #55

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
112 changes: 52 additions & 60 deletions test/ocr-engine-test.js
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
import { readFile } from "node:fs/promises";
import assert from "node:assert/strict";
import { readFile } from "node:fs/promises";
import { after, before, describe, it } from "node:test";

import sharp from "sharp";
Expand Down Expand Up @@ -173,43 +173,39 @@ describe("OCREngine", () => {
assert.strictEqual(dpi, varValue);
});

it(
"extracts bounding boxes from image",
{ timeout: 2_000 },
async function () {
const imageData = await loadImage(resolve("./small-test-page.jpg"));
ocr.loadImage(imageData);

// nb. The number of boxes returned here is slightly different than the
// test below which reads text boxes. This is because `getBoundingBoxes`
// performs a faster/simpler analysis and `getTextBoxes` triggers the more
// expensive LSTM-based analysis.
const wordBoxes = ocr.getBoundingBoxes("word");
assert.strictEqual(wordBoxes.length, 153);
it("extracts bounding boxes from image", { timeout: 2_000 }, async () => {
const imageData = await loadImage(resolve("./small-test-page.jpg"));
ocr.loadImage(imageData);

for (let box of wordBoxes) {
const { rect } = box;
// nb. The number of boxes returned here is slightly different than the
// test below which reads text boxes. This is because `getBoundingBoxes`
// performs a faster/simpler analysis and `getTextBoxes` triggers the more
// expensive LSTM-based analysis.
const wordBoxes = ocr.getBoundingBoxes("word");
assert.strictEqual(wordBoxes.length, 153);

assert.strictEqual(typeof rect.left, "number");
assert.strictEqual(typeof rect.right, "number");
assert.strictEqual(typeof rect.top, "number");
assert.strictEqual(typeof rect.bottom, "number");
for (const box of wordBoxes) {
const { rect } = box;

assert.strict(rect.left >= 0 && rect.left <= imageData.width);
assert.strict(rect.right >= 0 && rect.right <= imageData.width);
assert.strict(rect.right > rect.left);
assert.strictEqual(typeof rect.left, "number");
assert.strictEqual(typeof rect.right, "number");
assert.strictEqual(typeof rect.top, "number");
assert.strictEqual(typeof rect.bottom, "number");

assert.strict(rect.top >= 0 && rect.top <= imageData.height);
assert.strict(rect.bottom >= 0 && rect.bottom <= imageData.height);
assert.strict(rect.bottom > rect.top);
}
assert.strict(rect.left >= 0 && rect.left <= imageData.width);
assert.strict(rect.right >= 0 && rect.right <= imageData.width);
assert.strict(rect.right > rect.left);

const lineBoxes = ocr.getBoundingBoxes("line");
assert.strictEqual(lineBoxes.length, 10);
},
);
assert.strict(rect.top >= 0 && rect.top <= imageData.height);
assert.strict(rect.bottom >= 0 && rect.bottom <= imageData.height);
assert.strict(rect.bottom > rect.top);
}

const lineBoxes = ocr.getBoundingBoxes("line");
assert.strictEqual(lineBoxes.length, 10);
});

it("can extract bounding boxes without a model loaded", async function () {
it("can extract bounding boxes without a model loaded", async () => {
const ocr = await createEngine({ loadModel: false });

const imageData = await loadImage(resolve("./small-test-page.jpg"));
Expand All @@ -219,18 +215,18 @@ describe("OCREngine", () => {
assert.strictEqual(wordBoxes.length, 153);
});

it("extracts text boxes from image", { timeout: 10_000 }, async function () {
it("extracts text boxes from image", { timeout: 10_000 }, async () => {
const imageData = await loadImage(resolve("./small-test-page.jpg"));
ocr.loadImage(imageData);

const wordBoxes = ocr.getTextBoxes("word");
assert.strictEqual(wordBoxes.length, 159);
assert.strictEqual(wordBoxes.at(0).text, "Image");
assert.strictEqual(wordBoxes.at(-1).text, "complexity.");
let meanLength = mean(wordBoxes.map((b) => b.text.length));
const meanLength = mean(wordBoxes.map((b) => b.text.length));
assert.strict(meanLength >= 4);
assert.strict(meanLength <= 8);
let meanConfidence = mean(wordBoxes.map((b) => b.confidence));
const meanConfidence = mean(wordBoxes.map((b) => b.confidence));
assert.strict(meanConfidence >= 0.9);

const lineBoxes = ocr.getTextBoxes("line");
Expand Down Expand Up @@ -270,7 +266,7 @@ describe("OCREngine", () => {
});
});

it("extracts layout flags from image", { timeout: 5_000 }, async function () {
it("extracts layout flags from image", { timeout: 5_000 }, async () => {
const imageData = await loadImage(resolve("./small-test-page.jpg"));
ocr.loadImage(imageData);

Expand Down Expand Up @@ -308,7 +304,7 @@ describe("OCREngine", () => {
]);
});

it("extracts text from image", { timeout: 5_000 }, async function () {
it("extracts text from image", { timeout: 5_000 }, async () => {
const imageData = await loadImage(resolve("./small-test-page.jpg"));
ocr.loadImage(imageData);

Expand All @@ -319,34 +315,30 @@ describe("OCREngine", () => {
"This thresholding is a critical step",
];

for (let phrase of expectedPhrases) {
for (const phrase of expectedPhrases) {
assert.strict(text.includes(phrase));
}
});

it(
"accepts emscripten module options",
{ timeout: 5_000 },
async function () {
let stderr = "";
const writeToStderr = (s) => {
stderr += s;
};
it("accepts emscripten module options", { timeout: 5_000 }, async () => {
let stderr = "";
const writeToStderr = (s) => {
stderr += s;
};

const ocr = await createEngine({
emscriptenModuleOptions: { printErr: writeToStderr },
});
const ocr = await createEngine({
emscriptenModuleOptions: { printErr: writeToStderr },
});

const imageData = await loadImage(resolve("./small-test-page.jpg"));
ocr.loadImage(imageData);
const imageData = await loadImage(resolve("./small-test-page.jpg"));
ocr.loadImage(imageData);

ocr.getText();
ocr.getText();

assert.strictEqual(stderr, "Estimating resolution as 171");
},
);
assert.strictEqual(stderr, "Estimating resolution as 171");
});

it("extracts hOCR from image", { timeout: 5_000 }, async function () {
it("extracts hOCR from image", { timeout: 5_000 }, async () => {
const imageData = await loadImage(resolve("./small-test-page.jpg"));
ocr.loadImage(imageData);

Expand All @@ -359,12 +351,12 @@ describe("OCREngine", () => {
`<span class='ocr_line' id='line_1_5' title="bbox 36 443 1026 462; baseline 0 -5; x_size 18; x_descenders 4; x_ascenders 3">`,
];

for (let phrase of expectedPhrases) {
for (const phrase of expectedPhrases) {
assert.strict(html.includes(phrase));
}
});

it("reports recognition progress", { timeout: 5_000 }, async function () {
it("reports recognition progress", { timeout: 5_000 }, async () => {
const imageData = await loadImage(resolve("./small-test-page.jpg"));
ocr.loadImage(imageData);

Expand All @@ -374,7 +366,7 @@ describe("OCREngine", () => {
});

assert.strict(progressSteps.length > 0);
for (let [i, progress] in progressSteps.entries()) {
for (const [i, progress] in progressSteps.entries()) {
assert.strict(progress >= 0);
assert.strict(progress <= 100);
if (i > 0) {
Expand All @@ -396,7 +388,7 @@ describe("OCREngine", () => {
it("can determine image orientation", async () => {
const imagePath = resolve("./small-test-page.jpg");

for (let rotation of [0, 90, 180, 270]) {
for (const rotation of [0, 90, 180, 270]) {
const image = await sharp(imagePath).ensureAlpha().rotate(rotation);

ocr.loadImage(await toImageData(image));
Expand Down
Loading