haotian/convert-the-model-to-rknn
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

添加测试rknn模型

Browse Source
This commit is contained in:
haotian 2025-08-15 16:00:37 +08:00
parent c6add9e1a8
commit 6331efe449
1 changed files with 516 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

516
010测试PaddleOCR_rknn模型.py Normal file
View File

@ -0,0 +1,516 @@
import cv2
import numpy as np
from rknn.api import RKNN
import math
import time
class PaddleOCRRKNN:
def __init__(self, det_model_path, rec_model_path, target='rk3588'):
"""
初始化RKNN推理器
Args:
det_model_path: 检测模型路径 (det.rknn)
rec_model_path: 识别模型路径 (rec.rknn)
target: 目标平台 ('rk3588', 'rk3566', 'rk3568', 'rv1103', 'rv1106', 'simulator')
"""
self.target = target
# 初始化检测模型
self.det_rknn = RKNN(verbose=True)
ret = self.det_rknn.load_rknn(det_model_path)
if ret != 0:
print(f'Load detection RKNN model failed! Error code: {ret}')
print('Please check if the model file exists and is valid')
exit(ret)
# 指定运行时平台
print(f'Initializing detection model on target: {target}')
ret = self.det_rknn.init_runtime(
target=target,
device_id=None,
perf_debug=False,
eval_mem=False,
async_mode=False,
core_mask=RKNN.NPU_CORE_AUTO # 自动选择NPU核心
)
if ret != 0:
print(f'Init detection runtime failed! Error code: {ret}')
print('Please check if you are running on the correct platform')
if target == 'simulator':
print('Note: Simulator mode requires different setup')
exit(ret)
print('Detection model loaded successfully')
# 初始化识别模型
self.rec_rknn = RKNN(verbose=True)
ret = self.rec_rknn.load_rknn(rec_model_path)
if ret != 0:
print(f'Load recognition RKNN model failed! Error code: {ret}')
print('Please check if the model file exists and is valid')
exit(ret)
# 指定运行时平台
print(f'Initializing recognition model on target: {target}')
ret = self.rec_rknn.init_runtime(
target=target,
device_id=None,
perf_debug=False,
eval_mem=False,
async_mode=False,
core_mask=RKNN.NPU_CORE_AUTO # 自动选择NPU核心
)
if ret != 0:
print(f'Init recognition runtime failed! Error code: {ret}')
print('Please check if you are running on the correct platform')
exit(ret)
print('Recognition model loaded successfully')
# 字符集（根据您的模型调整）
self.character = ['blank', '!', '"', '#', '$', '%', '&', "'", '(', ')', '*', '+',
',', '-', '.', '/', '0', '1', '2', '3', '4', '5', '6', '7', '8',
'9', ':', ';', '<', '=', '>', '?', '@', 'A', 'B', 'C', 'D', 'E',
'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R',
'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', '[', '\\', ']', '^', '_',
'`', 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l',
'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y',
'z', '{', '|', '}', '~'] + [chr(i) for i in range(19968, 40870)] # 中文字符
def preprocess_det(self, img, input_shape=(640, 640)):
"""
检测模型的图像预处理 - 适配RKNN
"""
h, w, _ = img.shape
target_h, target_w = input_shape
# 计算缩放比例 - 保持宽高比
ratio_h = target_h / h
ratio_w = target_w / w
ratio = min(ratio_h, ratio_w)
# 计算缩放后的尺寸
new_h = int(h * ratio)
new_w = int(w * ratio)
# 调整图像大小
resized_img = cv2.resize(img, (new_w, new_h))
# 创建目标尺寸的图像，用灰色填充
padded_img = np.ones((target_h, target_w, 3), dtype=np.uint8) * 114
# 计算居中位置
top = (target_h - new_h) // 2
left = (target_w - new_w) // 2
# 将缩放后的图像放到居中位置
padded_img[top:top+new_h, left:left+new_w] = resized_img
# RKNN通常需要uint8输入（如果量化时使用了uint8）
# 如果您的模型使用float输入，请取消下面的注释并注释掉return语句
# padded_img = padded_img.astype(np.float32)
# padded_img = (padded_img / 255.0 - [0.485, 0.456, 0.406]) / [0.229, 0.224, 0.225]
return padded_img, ratio, (top, left)
def preprocess_rec(self, img, input_shape=(320, 48)):
"""
识别模型的图像预处理 - 适配RKNN
"""
target_w, target_h = input_shape # 注意：宽度在前
h, w = img.shape[:2]
# 计算缩放比例，保持宽高比
ratio_h = target_h / h
ratio_w = target_w / w
ratio = min(ratio_h, ratio_w)
# 计算缩放后的尺寸
new_h = int(h * ratio)
new_w = int(w * ratio)
# 调整图像大小
resized_image = cv2.resize(img, (new_w, new_h))
# 创建目标尺寸的图像，用黑色填充
padded_image = np.zeros((target_h, target_w, 3), dtype=np.uint8)
# 将缩放后的图像放到左上角（识别模型通常左对齐）
padded_image[:new_h, :new_w] = resized_image
# RKNN通常需要uint8输入（如果量化时使用了uint8）
# 如果您的模型使用float输入，请取消下面的注释并注释掉return语句
# padded_image = padded_image.astype(np.float32)
# padded_image = (padded_image / 255.0 - [0.485, 0.456, 0.406]) / [0.229, 0.224, 0.225]
return padded_image
def postprocess_det_boxes(self, dt_boxes, ratio, padding_info, ori_shape):
"""
检测结果后处理 - 适配固定输入形状
"""
if dt_boxes is None or len(dt_boxes) == 0:
return None
ori_h, ori_w = ori_shape
top, left = padding_info
# 将坐标从模型输出空间转换回原图空间
dt_boxes[:, :, 0] = (dt_boxes[:, :, 0] - left) / ratio
dt_boxes[:, :, 1] = (dt_boxes[:, :, 1] - top) / ratio
# 裁剪到原图范围内
dt_boxes[:, :, 0] = np.clip(dt_boxes[:, :, 0], 0, ori_w)
dt_boxes[:, :, 1] = np.clip(dt_boxes[:, :, 1], 0, ori_h)
return dt_boxes
def boxes_from_bitmap(self, pred, bitmap, dest_width, dest_height, max_candidates=1000, box_thresh=0.6):
"""
从位图中提取文本框
"""
bitmap = bitmap.astype(np.uint8)
height, width = bitmap.shape
# 查找轮廓
contours, _ = cv2.findContours(bitmap, cv2.RETR_LIST, cv2.CHAIN_APPROX_SIMPLE)
num_contours = min(len(contours), max_candidates)
boxes = []
scores = []
for i in range(num_contours):
contour = contours[i]
points, sside = self.get_mini_boxes(contour)
if sside < 5:
continue
points = np.array(points)
score = self.box_score_fast(pred, points.reshape(-1, 2))
if box_thresh > score:
continue
# 扩展box
box = self.unclip(points, 1.5).reshape(-1, 1, 2)
box, sside = self.get_mini_boxes(box)
if sside < 5 + 2:
continue
box = np.array(box)
box[:, 0] = np.clip(box[:, 0] / width * dest_width, 0, dest_width)
box[:, 1] = np.clip(box[:, 1] / height * dest_height, 0, dest_height)
boxes.append(box.astype(np.int16))
scores.append(score)
return np.array(boxes), scores
def get_mini_boxes(self, contour):
"""获取最小外接矩形"""
bounding_box = cv2.minAreaRect(contour)
points = sorted(list(cv2.boxPoints(bounding_box)), key=lambda x: x[0])
index_1, index_2, index_3, index_4 = 0, 1, 2, 3
if points[1][1] > points[0][1]:
index_1 = 0
index_4 = 1
else:
index_1 = 1
index_4 = 0
if points[3][1] > points[2][1]:
index_2 = 2
index_3 = 3
else:
index_2 = 3
index_3 = 2
box = [points[index_1], points[index_2], points[index_3], points[index_4]]
return box, min(bounding_box[1])
def box_score_fast(self, bitmap, _box):
"""快速计算box得分"""
h, w = bitmap.shape[:2]
box = _box.copy()
xmin = np.clip(np.floor(box[:, 0].min()).astype(int), 0, w - 1)
xmax = np.clip(np.ceil(box[:, 0].max()).astype(int), 0, w - 1)
ymin = np.clip(np.floor(box[:, 1].min()).astype(int), 0, h - 1)
ymax = np.clip(np.ceil(box[:, 1].max()).astype(int), 0, h - 1)
mask = np.zeros((ymax - ymin + 1, xmax - xmin + 1), dtype=np.uint8)
box[:, 0] = box[:, 0] - xmin
box[:, 1] = box[:, 1] - ymin
cv2.fillPoly(mask, box.reshape(1, -1, 2).astype(np.int32), 1)
return cv2.mean(bitmap[ymin:ymax + 1, xmin:xmax + 1], mask)[0]
def unclip(self, box, unclip_ratio):
"""扩展文本框"""
try:
from shapely.geometry import Polygon
import pyclipper
poly = Polygon(box)
distance = poly.area * unclip_ratio / poly.length
offset = pyclipper.PyclipperOffset()
offset.AddPath(box, pyclipper.JT_ROUND, pyclipper.ET_CLOSEDPOLYGON)
expanded = offset.Execute(distance)
if len(expanded) == 0:
return box
else:
return np.array(expanded[0])
except:
# 如果shapely/pyclipper不可用，使用简单的膨胀
return box
def decode_rec_result(self, preds_prob):
"""
解码识别结果
"""
preds_idx = np.argmax(preds_prob, axis=1)
preds_prob_max = np.max(preds_prob, axis=1)
# CTC解码
last_idx = 0
preds_text = []
preds_conf = []
for i, idx in enumerate(preds_idx):
if idx != last_idx and idx != 0: # 0是blank
if idx < len(self.character):
preds_text.append(self.character[idx])
preds_conf.append(preds_prob_max[i])
last_idx = idx
text = ''.join(preds_text)
conf = np.mean(preds_conf) if preds_conf else 0.0
return text, conf
def detect_text(self, image):
"""
文本检测 - RKNN版本
"""
ori_h, ori_w = image.shape[:2]
# 预处理
det_img, ratio, padding_info = self.preprocess_det(image)
# RKNN推理
start_time = time.time()
outputs = self.det_rknn.inference(inputs=[det_img])
det_time = time.time() - start_time
print(f"Detection inference time: {det_time:.3f}s")
# 获取输出 (通常是第一个输出)
det_output = outputs[0]
# 确保输出形状正确
if len(det_output.shape) == 4:
mask = det_output[0, 0, :, :]
else:
mask = det_output[0, :, :] if len(det_output.shape) == 3 else det_output
# 后处理
threshold = 0.3
bitmap = (mask > threshold).astype(np.uint8) * 255
# 从位图中提取文本框（坐标是在640x640空间中的）
boxes, scores = self.boxes_from_bitmap(mask, bitmap, 640, 640)
# 将坐标转换回原图空间
if len(boxes) > 0:
boxes = self.postprocess_det_boxes(boxes, ratio, padding_info, (ori_h, ori_w))
return boxes, scores
def recognize_text(self, image):
"""
文本识别 - RKNN版本
"""
# 预处理
rec_img = self.preprocess_rec(image)
# RKNN推理
start_time = time.time()
outputs = self.rec_rknn.inference(inputs=[rec_img])
rec_time = time.time() - start_time
print(f"Recognition inference time: {rec_time:.3f}s")
# 获取输出
rec_output = outputs[0]
# 确保输出维度正确
if len(rec_output.shape) == 3:
rec_result = rec_output[0] # 移除batch维度
else:
rec_result = rec_output
# 解码
text, conf = self.decode_rec_result(rec_result)
return text, conf
def get_rotate_crop_image(self, img, points):
"""
根据四个点坐标裁剪并矫正图像
"""
img_crop_width = int(
max(
np.linalg.norm(points[0] - points[1]),
np.linalg.norm(points[2] - points[3])))
img_crop_height = int(
max(
np.linalg.norm(points[0] - points[3]),
np.linalg.norm(points[1] - points[2])))
pts_std = np.float32([[0, 0], [img_crop_width, 0],
[img_crop_width, img_crop_height],
[0, img_crop_height]])
M = cv2.getPerspectiveTransform(points, pts_std)
dst_img = cv2.warpPerspective(
img,
M, (img_crop_width, img_crop_height),
borderMode=cv2.BORDER_REPLICATE,
flags=cv2.INTER_CUBIC)
dst_img_height, dst_img_width = dst_img.shape[0:2]
if dst_img_height * 1.0 / dst_img_width >= 1.5:
dst_img = np.rot90(dst_img)
return dst_img
def ocr(self, image_path):
"""
完整的OCR流程 - RKNN版本
"""
# 读取图像
image = cv2.imread(image_path)
if image is None:
print(f"Cannot read image: {image_path}")
return []
print(f"Processing image: {image.shape}")
# 1. 文本检测
dt_boxes, scores = self.detect_text(image)
if dt_boxes is None or len(dt_boxes) == 0:
print("No text boxes detected")
return []
print(f"Detected {len(dt_boxes)} text boxes")
# 2. 文本识别
ocr_results = []
for i, box in enumerate(dt_boxes):
# 裁剪文本区域
box_points = box.astype(np.float32)
crop_img = self.get_rotate_crop_image(image, box_points)
# 识别文本
text, conf = self.recognize_text(crop_img)
if conf > 0.5: # 置信度过滤
ocr_results.append({
'text': text,
'confidence': conf,
'box': box.tolist(),
'score': scores[i] if i < len(scores) else 0.0
})
print(f"Box {i}: {text} (conf: {conf:.3f})")
return ocr_results
def release(self):
"""
释放RKNN资源
"""
if hasattr(self, 'det_rknn'):
self.det_rknn.release()
if hasattr(self, 'rec_rknn'):
self.rec_rknn.release()
# 使用示例
def main():
# 初始化OCR - 可以指定不同的target平台
print("Initializing PaddleOCR RKNN...")
# 根据您的硬件平台选择：
# 'rk3588' - RK3588 (默认)
# 'rk3566' - RK3566
# 'rk3568' - RK3568
# 'rv1103' - RV1103
# 'rv1106' - RV1106
# 'simulator' - 仿真器模式（用于开发调试）
try:
ocr = PaddleOCRRKNN('/home/admin-root/haotian/康达瑞贝斯机器狗/det_shape.rknn', '/home/admin-root/haotian/康达瑞贝斯机器狗/rec_shape.rknn', target='rk3588')
except Exception as e:
print(f"Failed to initialize RKNN models: {e}")
print("Trying alternative initialization...")
# 如果失败，尝试其他选项
try:
ocr = PaddleOCRRKNN('det.rknn', 'rec.rknn', target='simulator')
except Exception as e2:
print(f"Alternative initialization also failed: {e2}")
return
try:
# 执行OCR
image_path = '/home/admin-root/haotian/康达瑞贝斯机器狗/data_image/001读表图片/3aee64cc1f90d93a5a45979f7b17cb4b_frame_001460.jpg'
# 检查图像文件是否存在
import os
if not os.path.exists(image_path):
print(f"Image file not found: {image_path}")
print("Please provide a valid image path")
return
results = ocr.ocr(image_path)
# 打印结果
print(f"\n=== OCR Results ({len(results)} items) ===")
for i, result in enumerate(results):
print(f"\n[{i+1}]")
print(f"Text: {result['text']}")
print(f"Confidence: {result['confidence']:.3f}")
print(f"Detection Score: {result['score']:.3f}")
print(f"Box: {result['box']}")
# 可视化结果
if results:
visualize_results(image_path, results)
else:
print("No text detected in the image")
except Exception as e:
print(f"OCR processing failed: {e}")
finally:
# 释放资源
ocr.release()
print("RKNN resources released")
def visualize_results(image_path, results):
"""
可视化OCR结果
"""
image = cv2.imread(image_path)
for i, result in enumerate(results):
box = np.array(result['box'], dtype=np.int32)
cv2.polylines(image, [box], True, (0, 255, 0), 2)
# 在框上方显示文本和序号
text_display = f"{i+1}: {result['text'][:20]}..."
cv2.putText(image, text_display,
(box[0][0], box[0][1] - 10),
cv2.FONT_HERSHEY_SIMPLEX, 0.7, (255, 0, 0), 2)
# 保存结果图像
output_path = 'ocr_result_rknn.jpg'
cv2.imwrite(output_path, image)
print(f"Result image saved to: {output_path}")
if __name__ == "__main__":
main()